26 |
32 | 21 | This specification does neat stuff. 22 |
23 |26 | This is an unofficial proposal. 27 |
28 |33 | See ReSpec's user guide 34 | for how toget started! 35 |
36 |The following people contributed to the development of this document.
4 |This document provides samples of Pronunciation.
16 |This sample content has been developed by the W3C Accessible Platform Architecture Pronunciation Task Force to allow developers to 19 | evaluate and test their implementation of the proposed Attribute Model of SSML in HTML. The samples are presented in each of the two proposed 20 | approaches for representing SSML via attibutes. There are currently 9 test cases, each incorporating 21 | one or more of the SSML functions defined in the Pronunciation Technical Approach Document. Each test case includes an audio sample of the expected 22 | spoken presentation. 23 |
24 |Developers of read aloud tools, screen readers, and voice assistants should use these test cases to evaluate and test their implementation. Questions and test results 25 | should be filed as issues in the Pronunciation github repo.
26 |
26 |
32 | Developers of read aloud tools, screen readers, and voice assistants should use these test cases to evaluate and test their implementation. Questions and test results 17 | should be filed as issues in the Pronunciation github repo.
18 |According the 2010 US Census, the population of 90274 22 | increased to 25209 from 24976 over the past 10 years.
23 |Case 1: Audio rendering 24 |
Once upon a midnight dreary
26 |Case 2: Audio rendering 27 |
NaCL
29 |Case 3: Audio rendering 30 |
She said, "My name is Marie", to which, he replied, "I am Tom."
32 |Case 4: Audio rendering (no sample)
33 |Please use extreme caution..
35 |Case 5: Audio rendering 36 |
Take a deep breath, and exhale.
38 |Case 6: Audio rendering 39 |
The tortoise, said (slowly) "I am almost at the finish line."
41 |Case 7: Audio rendering 42 |
You will hear a brief chime when your time is up.
44 |Case 8: Audio rendering 45 |
47 | Once upon a midnight dreary
48 | ,
49 | while I pondered, weak
50 | and weary,
51 | Over many a quaint and curious volume of forgotton
52 | lore—
53 | While I nodded, nearly napping, suddenly there came a tapping,
54 |
55 | As of some one gently rapping,
56 |
57 | rapping at my chamber door.
58 |
59 |
60 | "'Tis some visitor,"
61 | I muttered,
62 | "tapping
63 | at my chamber door—
64 | Only this
65 | and nothing
66 | more."
67 |
Case 9: Audio rendering 69 | 70 | 71 | 72 | 73 | -------------------------------------------------------------------------------- /user-scenarios/respec-config.js: -------------------------------------------------------------------------------- 1 | var respecConfig = { 2 | // embed RDFa data in the output 3 | trace: true, 4 | doRDFa: '1.0', 5 | includePermalinks: true, 6 | permalinkEdge: true, 7 | permalinkHide: false, 8 | tocIntroductory: true, 9 | // specification status (e.g., WD, LC, NOTE, etc.). If in doubt use ED. 10 | specStatus: "ED", 11 | noRecTrack: true, 12 | //crEnd: "2012-04-30", 13 | //perEnd: "2013-07-23", 14 | //publishDate: "2013-08-22", 15 | //diffTool: "http://www.aptest.com/standards/htmldiff/htmldiff.pl", 16 | 17 | // the specifications short name, as in http://www.w3.org/TR/short-name/ 18 | shortName: "pronunciation-user-scenarios", 19 | 20 | 21 | // if you wish the publication date to be other than today, set this 22 | //publishDate: "2017-05-09", 23 | copyrightStart: "2018", 24 | license: "w3c-software", 25 | 26 | // if there is a previously published draft, uncomment this and set its YYYY-MM-DD date 27 | // and its maturity status 28 | //previousPublishDate: "", 29 | //previousMaturity: "", 30 | //prevRecURI: "", 31 | //previousDiffURI: "", 32 | 33 | // if there a publicly available Editors Draft, this is the link 34 | edDraftURI: "https://w3c.github.io/pronunciation/user-scenarios", 35 | 36 | // if this is a LCWD, uncomment and set the end of its review period 37 | // lcEnd: "2012-02-21", 38 | 39 | // editors, add as many as you like 40 | // only "name" is required 41 | editors: [ 42 | { 43 | name: "Irfan Ali", 44 | url: 'https://www.w3.org/users/98332', 45 | mailto: "irfan.ali@blackRock.com", 46 | company: "BlackRock", 47 | companyURI: "https://www.blackrock.com/", 48 | w3cid: 98332 49 | }, 50 | 51 | { 52 | name: "Sam Kanta", 53 | mailto: "sam.kanta@thesustainablechange.com" 54 | }, 55 | { 56 | name: "Christine Loew", 57 | company: "College Board", 58 | mailto: "cloew@collegeboard.org" 59 | }, 60 | { 61 | name: "Paul Grenier", 62 | company: "Deque System", 63 | mailto: "paul.grenier@deque.com" 64 | 65 | }, 66 | { 67 | name: "Roy Ran", 68 | url: 'https://www.w3.org', 69 | mailto: "ran@w3.org", 70 | company: "W3C", 71 | companyURI: "http://www.w3.org", 72 | w3cid: 100586 73 | } 74 | ], 75 | 76 | // authors, add as many as you like. 77 | // This is optional, uncomment if you have authors as well as editors. 78 | // only "name" is required. Same format as editors. 79 | 80 | //authors: [ 81 | // { name: "Your Name", url: "http://example.org/", 82 | // company: "Your Company", companyURI: "http://example.com/" }, 83 | //], 84 | 85 | /* 86 | alternateFormats: [ 87 | { uri: 'aria-diff.html', label: "Diff from Previous Recommendation" } , 88 | { uri: 'aria.ps', label: "PostScript version" }, 89 | { uri: 'aria.pdf', label: "PDF version" } 90 | ], 91 | */ 92 | 93 | // errata: 'http://www.w3.org/2010/02/rdfa/errata.html', 94 | 95 | // name of the WG 96 | //wg: "Accessible Platform Architectures Working Group", 97 | 98 | // URI of the public WG page 99 | //wgURI: "https://www.w3.org/WAI/APA/", 100 | 101 | // name (with the @w3c.org) of the public mailing to which comments are due 102 | wgPublicList: "public-pronunciation", 103 | 104 | // URI of the patent status for this WG, for Rec-track documents 105 | // !!!! IMPORTANT !!!! 106 | // This is important for Rec-track documents, do not copy a patent URI from a random 107 | // document unless you know what you're doing. If in doubt ask your friendly neighbourhood 108 | // Team Contact. 109 | //wgPatentURI: "https://www.w3.org/2004/01/pp-impl/83907/status", 110 | //maxTocLevel: 2, 111 | 112 | //group: 83907 113 | group: "apa", 114 | github: "https://github.com/w3c/pronunciation/" 115 | 116 | // Spec URLs 117 | 118 | 119 | }; 120 | -------------------------------------------------------------------------------- /gap-analysis/respec-config.js: -------------------------------------------------------------------------------- 1 | var respecConfig = { 2 | // embed RDFa data in the output 3 | trace: true, 4 | doRDFa: '1.0', 5 | includePermalinks: true, 6 | permalinkEdge: true, 7 | permalinkHide: false, 8 | tocIntroductory: true, 9 | // specification status (e.g., WD, LC, NOTE, etc.). If in doubt use ED. 10 | specStatus: "ED", 11 | noRecTrack: true, 12 | //crEnd: "2012-04-30", 13 | //perEnd: "2013-07-23", 14 | //publishDate: "2013-08-22", 15 | //diffTool: "http://www.aptest.com/standards/htmldiff/htmldiff.pl", 16 | 17 | // the specifications short name, as in http://www.w3.org/TR/short-name/ 18 | shortName: "pronunciation-gap-analysis", 19 | 20 | 21 | // if you wish the publication date to be other than today, set this 22 | //publishDate: "2017-05-09", 23 | copyrightStart: "2018", 24 | license: "w3c-software", 25 | 26 | // if there is a previously published draft, uncomment this and set its YYYY-MM-DD date 27 | // and its maturity status 28 | //previousPublishDate: "", 29 | //previousMaturity: "", 30 | //prevRecURI: "", 31 | //previousDiffURI: "", 32 | 33 | // if there a publicly available Editors Draft, this is the link 34 | edDraftURI: "https://w3c.github.io/pronunciation/gap-analysis", 35 | 36 | // if this is a LCWD, uncomment and set the end of its review period 37 | // lcEnd: "2012-02-21", 38 | 39 | // editors, add as many as you like 40 | // only "name" is required 41 | editors: [ 42 | { 43 | name: "Markku Hakkinen", 44 | url: 'https://www.w3.org/users/35712', 45 | mailto: "mhakkinen@ets.org", 46 | company: "Educational Testing Service", 47 | companyURI: "https://www.ets.org/", 48 | w3cid: 35712 49 | }, 50 | { 51 | name: "Steve Noble", 52 | url: 'https://www.w3.org/users/112867', 53 | mailto: "steve.noble@pearson.com", 54 | company: "Pearson", 55 | companyURI: "https://www.pearson.com/", 56 | w3cid: 98332 57 | }, 58 | { 59 | name: "Irfan Ali", 60 | url: 'https://www.w3.org/users/98332', 61 | mailto: "irfan.ali@blackrock.com", 62 | company: "BlackRock", 63 | companyURI: "https://www.blackrock.com", 64 | w3cid: 98332 65 | }, 66 | 67 | 68 | { 69 | name: "Roy Ran", 70 | url: 'https://www.w3.org', 71 | mailto: "ran@w3.org", 72 | company: "W3C", 73 | companyURI: "http://www.w3.org", 74 | w3cid: 100586 75 | } 76 | ], 77 | 78 | // authors, add as many as you like. 79 | // This is optional, uncomment if you have authors as well as editors. 80 | // only "name" is required. Same format as editors. 81 | 82 | //authors: [ 83 | // { name: "Your Name", url: "http://example.org/", 84 | // company: "Your Company", companyURI: "http://example.com/" }, 85 | //], 86 | 87 | /* 88 | alternateFormats: [ 89 | { uri: 'aria-diff.html', label: "Diff from Previous Recommendation" } , 90 | { uri: 'aria.ps', label: "PostScript version" }, 91 | { uri: 'aria.pdf', label: "PDF version" } 92 | ], 93 | */ 94 | 95 | // errata: 'http://www.w3.org/2010/02/rdfa/errata.html', 96 | 97 | // name of the WG 98 | wg: "Accessible Platform Architectures Working Group", 99 | 100 | // URI of the public WG page 101 | wgURI: "https://www.w3.org/WAI/APA/", 102 | 103 | // name (with the @w3c.org) of the public mailing to which comments are due 104 | wgPublicList: "public-pronunciation", 105 | 106 | // URI of the patent status for this WG, for Rec-track documents 107 | // !!!! IMPORTANT !!!! 108 | // This is important for Rec-track documents, do not copy a patent URI from a random 109 | // document unless you know what you're doing. If in doubt ask your friendly neighbourhood 110 | // Team Contact. 111 | wgPatentURI: "https://www.w3.org/2004/01/pp-impl/83907/status", 112 | //maxTocLevel: 2, 113 | 114 | 115 | 116 | // Spec URLs 117 | 118 | 119 | }; 120 | -------------------------------------------------------------------------------- /samples/w3cptf-multiattr-tests.html: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 |
6 |Developers of read aloud tools, screen readers, and voice assistants should use these test cases to evaluate and test their implementation. Questions and test results 18 | should be filed as issues in the Pronunciation github repo.
19 |According the 2010 US Census, the population of 90274 22 | increased to 25209 from 24976 over the past 10 years.
23 |Case 1: Audio rendering 24 |
Once upon a midnight dreary
26 |Case 2: Audio rendering 27 |
NaCL
29 |Case 3: Audio rendering 30 |
She said, "My name is Marie", to which, he replied, I am Tom."
32 |Case 4: Audio rendering (no sample)
33 |Please use extreme caution..
35 |Case 5: Audio rendering 36 |
ake a deep breath, and exhale.
38 |Case 6: Audio rendering 39 |
The tortoise, said (slowly) " 41 | I am almost at the finish line."
42 |Case 7: Audio rendering 43 |
You will hear a brief chime when your time is up.
45 |Case 8: Audio rendering 46 |
48 | Once upon a midnight dreary
49 | ,
50 | while I pondered, weak
51 | and weary,
52 | Over many a quaint and curious volume of forgotton lore—
54 | While I nodded, nearly napping, suddenly there came a tapping,
55 |
56 | As of some one gently rapping,
57 |
58 | rapping at my chamber door.
59 |
60 |
61 |
62 | "'Tis some visitor,"
63 | I muttered,
"tapping at my chamber door—
66 | Only this and nothing
67 |
more."
68 |
Case 9: Audio rendering 71 | 72 | 73 | 74 | 75 | -------------------------------------------------------------------------------- /gap-analysis_and_use-case/respec-config.js: -------------------------------------------------------------------------------- 1 | var respecConfig = { 2 | // embed RDFa data in the output 3 | trace: true, 4 | doRDFa: '1.0', 5 | includePermalinks: true, 6 | permalinkEdge: true, 7 | permalinkHide: false, 8 | tocIntroductory: true, 9 | // specification status (e.g., WD, LC, NOTE, etc.). If in doubt use ED. 10 | specStatus: "ED", 11 | noRecTrack: true, 12 | //crEnd: "2012-04-30", 13 | //perEnd: "2013-07-23", 14 | //publishDate: "2013-08-22", 15 | //diffTool: "http://www.aptest.com/standards/htmldiff/htmldiff.pl", 16 | 17 | // the specifications short name, as in http://www.w3.org/TR/short-name/ 18 | shortName: "pronunciation-gap-analysis-and-use-cases", 19 | 20 | 21 | // if you wish the publication date to be other than today, set this 22 | //publishDate: "2017-05-09", 23 | copyrightStart: "2020", 24 | 25 | 26 | // if there is a previously published draft, uncomment this and set its YYYY-MM-DD date 27 | // and its maturity status 28 | //previousPublishDate: "", 29 | //previousMaturity: "", 30 | //prevRecURI: "", 31 | //previousDiffURI: "", 32 | 33 | // if there a publicly available Editors Draft, this is the link 34 | edDraftURI: "https://w3c.github.io/pronunciation/gap-analysis_and_use-case", 35 | 36 | // if this is a LCWD, uncomment and set the end of its review period 37 | // lcEnd: "2012-02-21", 38 | 39 | // editors, add as many as you like 40 | // only "name" is required 41 | editors: [ 42 | { 43 | name: "Markku Hakkinen", 44 | url: 'https://www.w3.org/users/35712', 45 | mailto: "mhakkinen@ets.org", 46 | company: "Educational Testing Service", 47 | companyURI: "https://www.ets.org/", 48 | w3cid: 35712 49 | }, 50 | { 51 | name: "Steve Noble", 52 | url: 'https://www.w3.org/users/112867', 53 | mailto: "steve.noble@pearson.com", 54 | company: "Pearson", 55 | companyURI: "https://www.pearson.com/", 56 | w3cid: 98332 57 | }, 58 | { 59 | name: "Dee Dyer", 60 | mailto: "ddyer@ets.org", 61 | company: "Educational Testing Service", 62 | companyURI: "https://www.ets.org/", 63 | w3cid: 00000 64 | 65 | }, 66 | { 67 | name: "Irfan Ali", 68 | url: 'https://www.w3.org/users/98332', 69 | mailto: "iali@ets.org", 70 | company: "Educational Testing Service", 71 | companyURI: "https://www.ets.org/", 72 | w3cid: 98332 73 | }, 74 | { 75 | name: "Paul Grenier", 76 | mailto: "pgrenier@gmail.com" 77 | 78 | }, 79 | 80 | { 81 | name: "Roy Ran", 82 | url: 'https://www.w3.org', 83 | mailto: "ran@w3.org", 84 | company: "W3C", 85 | companyURI: "http://www.w3.org", 86 | w3cid: 100586 87 | } 88 | ], 89 | 90 | // authors, add as many as you like. 91 | // This is optional, uncomment if you have authors as well as editors. 92 | // only "name" is required. Same format as editors. 93 | 94 | //authors: [ 95 | // { name: "Your Name", url: "http://example.org/", 96 | // company: "Your Company", companyURI: "http://example.com/" }, 97 | //], 98 | 99 | /* 100 | alternateFormats: [ 101 | { uri: 'aria-diff.html', label: "Diff from Previous Recommendation" } , 102 | { uri: 'aria.ps', label: "PostScript version" }, 103 | { uri: 'aria.pdf', label: "PDF version" } 104 | ], 105 | */ 106 | 107 | // errata: 'http://www.w3.org/2010/02/rdfa/errata.html', 108 | 109 | // name of the WG 110 | //wg: "Accessible Platform Architectures Working Group", 111 | 112 | // URI of the public WG page 113 | //wgURI: "https://www.w3.org/WAI/APA/", 114 | 115 | // name (with the @w3c.org) of the public mailing to which comments are due 116 | //wgPublicList: "public-pronunciation", 117 | 118 | // URI of the patent status for this WG, for Rec-track documents 119 | // !!!! IMPORTANT !!!! 120 | // This is important for Rec-track documents, do not copy a patent URI from a random 121 | // document unless you know what you're doing. If in doubt ask your friendly neighbourhood 122 | // Team Contact. 123 | //wgPatentURI: "https://www.w3.org/2004/01/pp-impl/83907/status", 124 | //maxTocLevel: 2, 125 | 126 | group: "apa", 127 | github: "https://github.com/w3c/pronunciation/" 128 | 129 | // Spec URLs 130 | 131 | 132 | }; 133 | -------------------------------------------------------------------------------- /technical-approach/ssml-json-schema-w3cptf.json: -------------------------------------------------------------------------------- 1 | { 2 | "$schema": "http://json-schema.org/draft-07/schema#", 3 | "$id": "http://ets-research.org/ia11ylab/ia/json/ssml-json-schema-w3cptf.json", 4 | "title": "SSML as a single attribute for inclusion in HTML", 5 | "description": "JSON structure representing each SSML element as a JSON object. The SSML properties are dervived from https://www.w3.org/TR/speech-synthesis11/. Several elements are excluded: mark, speak, p, w and the desc attribute. Author: M. Hakkinen - ETS", 6 | "type": "object", 7 | "properties": { 8 | "say-as": { 9 | "description": "The unique identifier for a product", 10 | "type": "object", 11 | "properties": { 12 | "interpret-as": { "type": "string", 13 | "enum": ["date","time","telephone","characters","cardinal","ordinal"]}, 14 | "format": { "type": "string" }, 15 | "detail": {"type": "string"} 16 | } 17 | }, 18 | "phoneme": { 19 | "description": "The Phoneme Function", 20 | "type": "object", 21 | "properties": { 22 | "ph": { "type": "string"}, 23 | "alphabet": {"type": "string", 24 | "enum": ["ipa", "x-sampa"]}} 25 | }, 26 | "sub": { 27 | "description": "sub function", 28 | "type": "object", 29 | "properties": { 30 | "alias": {"type":"string"} 31 | } 32 | }, 33 | "voice":{ 34 | "description": "voice function", 35 | "type":"object", 36 | "properties": { 37 | "gender": {"type":"string", 38 | "enum": ["female","male","neutral"]}, 39 | "age": {"type":"integer"}, 40 | "variant":{"type":"string"}, 41 | "name": {"type":"string"}, 42 | "languages": {"type":"string"} 43 | } 44 | }, 45 | "emphasis":{ 46 | "description": "speech emphasis level", 47 | "type":"object", 48 | "properties": { 49 | "level": {"type":"string", 50 | "enum": ["none","x-weak","weak","medium","strong","x-strong"]}, 51 | "time": {"type":"string", 52 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"} 53 | } 54 | }, 55 | "prosody": { 56 | "description": "speech prosody", 57 | "type":"object", 58 | "properties": { 59 | "pitch": {"type":"string", 60 | "pattern":"^x-low|low|medium|high|x-high|default|(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)Hz)$" 61 | }, 62 | "contour": {"type":"string"}, 63 | "range": {"type":"string", 64 | "pattern":"^x-low|low|medium|high|x-high|default|(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)Hz)$" 65 | }, 66 | "rate": {"type":"string", 67 | "pattern":"^x-slow|slow|medium|fast|x-xfast|default|(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)%)$"}, 68 | "duration": {"type": "string", 69 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"}, 70 | "volume": {"type":"string", 71 | "pattern":"^silent|x-soft|soft|medium|loud|x-loud|default|(+|-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)dB)$"} 72 | } 73 | }, 74 | "break": { 75 | "description": "break - insert a timed pause", 76 | "type":"object", 77 | "properties": { 78 | "strength": {"type":"string", 79 | "enum": ["none","x-weak","weak","medium","strong","x-strong"]}, 80 | "time": {"type":"string", 81 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"} 82 | } 83 | }, 84 | "audio": { 85 | "description":"audio element used to insert audio file into speech stream", 86 | "type":"object", 87 | "properties":{ 88 | "src": {"type":"uri"}, 89 | "fetchtimeout":{"type":"string", 90 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"}, 91 | "fetchint":{"type":"string", 92 | "enum": ["safe","prefetch"]}, 93 | "maxage":{"type":"string"}, 94 | "maxstale":{"type":"string"}, 95 | "clipBegin":{"type": "string", 96 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$" 97 | 98 | }, 99 | "clipEnd":{"type": "string", 100 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$" 101 | 102 | }, 103 | "repeatCount":{"type":"integer" 104 | }, 105 | "repeatDur":{"type": "string", 106 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$" 107 | 108 | }, 109 | "soundLevel":{"type":"string", 110 | "pattern":"^(+|-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)dB)$"}, 111 | 112 | "speed":{ 113 | "type":"string", 114 | "pattern":"^((0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)%)$" 115 | } 116 | 117 | } 118 | } 119 | } 120 | } 121 | 122 | -------------------------------------------------------------------------------- /technical-approach/appendixJSON.html: -------------------------------------------------------------------------------- 1 | 2 | 3 |
4 |53 | The Pronunciation Task Force develops specifications for hypertext markup language (HTML) author control of 54 | text-to-speech (TTS) presentation. 55 |
56 |The JSON schema defines the specific SSML functions, properties, and values recommended for implementation in 61 | this proposal.
62 | 63 |
65 |
66 | {
67 | "$schema": "http://json-schema.org/draft-07/schema#",
68 | "$id": "http://ets-research.org/ia11ylab/ia/json/ssml-json-schema-w3cptf.json",
69 | "title": "SSML as a single attribute for inclusion in HTML",
70 | "description": "JSON structure representing each SSML element as a JSON object. The SSML properties are dervived
71 | from https://www.w3.org/TR/speech-synthesis11/. Several elements are excluded: mark, speak, p, w and the desc attribute.
72 | Author: M. Hakkinen - ETS",
73 | "type": "object",
74 | "properties": {
75 | "say-as": {
76 | "description": "The unique identifier for a product",
77 | "type": "object",
78 | "properties": {
79 | "interpret-as": { "type": "string",
80 | "enum": ["date","time","telephone","characters","cardinal","ordinal"]},
81 | "format": { "type": "string" },
82 | "detail": {"type": "string"}
83 | }
84 | },
85 | "phoneme": {
86 | "description": "The Phoneme Function",
87 | "type": "object",
88 | "properties": {
89 | "ph": { "type": "string"},
90 | "alphabet": {"type": "string", "enum": ["ipa", "x-sampa"]}}
91 | },
92 | "sub": {
93 | "description": "sub function", "type": "object",
94 | "properties": {
95 | "alias": {"type":"string"}}
96 | },
97 | "voice":{"description": "voice function", "type":"object",
98 | "properties": {
99 | "gender": {"type":"string",
100 | "enum": ["female","male","neutral"]},
101 | "age": {"type":"integer"},
102 | "variant":{"type":"string"},
103 | "name": {"type":"string"},
104 | "languages": {"type":"string"}
105 | }
106 | },
107 | "emphasis":{
108 | "description": "speech emphasis level",
109 | "type":"object",
110 | "properties": {
111 | "level": {"type":"string",
112 | "enum": ["none","x-weak","weak","medium","strong","x-strong"]},
113 | "time": {"type":"string",
114 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"}
115 | }
116 | },
117 | "prosody": {
118 | "description": "speech prosody",
119 | "type":"object",
120 | "properties": {
121 | "pitch": {"type":"string",
122 | "pattern":"^x-low|low|medium|high|x-high|default|(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)Hz)$"},
123 | "contour": {"type":"string"},
124 | "range": {"type":"string",
125 | "pattern":"^x-low|low|medium|high|x-high|default|(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)Hz)$"},
126 | "rate": {"type":"string",
127 | "pattern":"^x-slow|slow|medium|fast|x-xfast|default|(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)%)$"},
128 | "duration": {"type": "string",
129 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"},
130 | "volume": {"type":"string",
131 | "pattern":"^silent|x-soft|soft|medium|loud|x-loud|default|(+|-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)dB)$"}
132 | }
133 | },
134 | "break": {
135 | "description": "break - insert a timed pause",
136 | "type":"object",
137 | "properties": {
138 | "strength": {"type":"string",
139 | "enum": ["none","x-weak","weak","medium","strong","x-strong"]},
140 | "time": {"type":"string",
141 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"}
142 | }
143 | },
144 | "audio": {
145 | "description":"audio element used to insert audio file into speech stream",
146 | "type":"object",
147 | "properties":{
148 | "src": {"type":"uri"},
149 | "fetchtimeout":{"type":"string",
150 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"},
151 | "fetchint":{"type":"string",
152 | "enum": ["safe","prefetch"]},
153 | "maxage":{"type":"string"},
154 | "maxstale":{"type":"string"},
155 | "clipBegin":{"type": "string",
156 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"},
157 | "clipEnd":{"type": "string",
158 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"},
159 | "repeatCount":{"type":"integer"
160 | "repeatDur":{"type": "string",
161 | "pattern":"^(-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)ms|s)$"},
162 | "soundLevel":{"type":"string",
163 | "pattern":"^(+|-?(0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)dB)$"},
164 | "speed":{
165 | "type":"string",
166 | "pattern":"^((0|[1-9]\\d*)?(\\.\\d+)?(?<=\\d)%)$"}
167 | }
168 | }
169 | }
170 | }
171 |
172 |
173 | The objective of the Pronunciation Task Force is to develop normative specifications and best practices guidance collaborating with other W3C groups as appropriate, to provide for proper pronunciation in HTML content when using text to speech (TTS) synthesis. This document defines a standard mechanism to allow 17 | content authors to include spoken presentation guidance in HTML content.
18 | 19 |Accurate, consistent pronunciation and presentation of content spoken by text-to-speech (TTS) synthesis is an essential requirement in education, 27 | communication, entertainment, and other domains. 28 | From helping to teach spelling and pronunciation in different languages, to reading learning materials or new stories, 29 | TTS has become a vital technology for providing access to digital content on the web, through mobile devices, and now via voice-based assistants. 30 | Organizations such as educational publishers and assessment vendors are looking for a standards-based solution to 31 | enable authoring of spoken presentation guidance in HTML which can then be consumed by assistive technologies (AT) and other 32 | applications that utilize text to speech synthesis (TTS) for rendering of content. 33 | Historically, efforts at standardization (e.g. SSML or CSS Speech) have not led to broad adoption of any standard by user agents, authors, or AT; 34 | what has arisen are a variety of non-interoperable approaches that meet specific needs for some applications. This explainer document presents the case for 35 | improving spoken presentation on the Web and how a standards-based approach can address the requirements.
36 | 37 | 38 |This is a proposal for a mechanism to allow content authors to include spoken presentation guidance in HTML content. Such guidance can be used by AT (including screen readers and read aloud tools) and voice assistants to control TTS synthesis. A key requirement is to ensure the spoken presentation content matches the author's intent and user expectations.
42 |The challenge is integrating pronunciation content into HTML so that it is easy to author, does not "break" content, and is straightforward for consumption by AT, voice assistants, and other tools that produce spoken presentation of content.
43 |Several classes of AT users depend upon spoken rendering of web content by TTS synthesis. In contexts such as education, there are specific expectations for accuracy of spoken presentation in terms of pronunciation, emphasis, prosody, pausing, etc.
47 |Correct pronunciation is also important in the context of language learning, where incorrect pronunciation can confuse learners.
48 |In practice, the ecosystem of devices used in classrooms is broad, and each vendor generally provides their own TTS engines for their platforms. Ensuring consistent spoken presentation across devices is a very real problem, and challenge. For many educational assessment vendors, the problem necessitates non-interoperable hacks to tune pronunciation and other presentation features, such as pausing, which itself can introduce new problems through inconsistent representation of text across speech and braille.
49 |It could be argued that continual advances in machine learning will improve the quality of synthesized speech, reducing the need for this proposal. Waiting for a robust solution that will likely still not fully address our needs is risky, especially when an authorable, declarative approach may be within reach (and wouldn't preclude or conflict with continual improvement in TTS technology).
50 |The current situation:
51 |With the growing consumer adoption of voice assistants, user expectations for high quality spoken presentation is growing. Google and Amazon both encourage application developers to utilize SSML to enhance the user experience on their platforms, yet Web content authors do not have the same opportunity to enhance the spoken presentation of their content.
58 |Finding a solution to this need can have broader benefit in allowing authors to create web content that presents a better user experience if the content is presented by voice assistants.
59 |<span> with ARIA attributes)According the 2010 US Census, the population
68 | of
According the 2010 US Census, the population 93 | of 90274 94 | increased to 25209 from 24976 over the past 10 years. 95 |
96 | ``` 97 | 98 | ## Open Questions 99 | 100 | 1. From the TAG/WHATWG perspective, what disadvantages/challenges have we missed with either approach? 101 | 2. Whichever approach makes sense from the web standards perspective, will/can it be adopted by assistive technologies? Particularly for screen readers, does it fit the accessibility API model? 102 | 103 | 104 | 105 | 106 | -------------------------------------------------------------------------------- /user-scenarios/draft.md: -------------------------------------------------------------------------------- 1 | # Introduction 2 | 3 | As part of the Accessible Platform Architectures (APA) Working Group, the Pronunciation Task Force (PTF) is a collaboration of subject matter experts working to identify and specify the optimal approach which can deliver reliably accurate pronunciation across browser and operating environments. With the introduction of the Kurzweil reading aid in 1976, to the more sophisticated synthetic speech currently used to assist communication as reading aids for the visually impaired and those with reading disabilities, the technology has multiple applications in education, communication, entertainment, etc. From helping to teach spelling and pronunciation in different languages, Text-to-Speech (TTS) has become a vital technology for providing access to digital content on the web and through mobile devices. 4 | 5 | The challenges that TTS presents include but are not limited to: the inability to accommodate regional variation and presentation of every phoneme present throughout the world; the incorrect determination by TTS of the pronunciation of content in context, and; the current inability to influence other pronunciation characteristics such as prosody and emphasis. 6 | 7 | # User Scenarios 8 | The purpose of developing user scenarios is to facilitate discussion and further requirements definition for pronunciation standards developed within the PTF prior to review of the APA. There are numerous interpretations of what form user scenarios adopt. Within the user experience research (UXR) body of practice, a user scenario is a written narrative related to the use of a service from the perspective of a user or user group. Importantly, the context of use is emphasized as is the desired outcome of use. There are potentially thousands of user scenarios for a technology such as TTS, however, the focus for the PTF is on the core scenarios that relate to the kinds of users who will engage with TTS. 9 | 10 | User scenarios, like Personas, represent a composite of real-world experiences. In the case of the PTF, the scenarios were derived from interviews of people who were end-consumers of TTS, as well as submitted narratives and industry examples from practitioners. There are several formats of scenarios. Several are general goal or task-oriented scenarios. Others elaborate on richer context, for example, educational assessment. 11 | 12 | The following user scenarios are organized on the three perspectives of TTS use derived from analysis of the qualitative data collected from the discovery work: 13 | 14 | + **End-Consumers of TTS:** Encompasses those with a visual disability or other need to have TTS operational when using assistive technologies (ATs). 15 | + **Digital Content Managers:** Addresses activities related to those responsible for producing content that needs to be accessible to ATs and W3C-WAI Guidelines. 16 | + **Software Engineers:** Includes developers and architects required to put TTS into an application or service. 17 | 18 | ## End-Consumers of TTS 19 | Ultimately, the quality and variation of TTS rendering by assistive technologies vary widely according to a user's context. The following user scenarios reinforce the necessity for accurate pronunciation from the perspective of those who consume digitally generated content. 20 | 21 | A. As a traveler who uses assistive technology (AT) with TTS to help navigate through websites, I need to hear arrival and destination codes pronounced accurately so I can select the desired travel itinerary. For example, a user with a visual impairment attempts to book a flight to Ottawa, Canada and so goes to a travel website. The user already knows the airport code and enters "YOW". The site produces the result in a drop-down list as "Ottawa, CA" but the AT does not pronounce the text accurately to help the user make the correct association between their data entry and the list item. 22 | 23 | B. As a test taker (tester) with a visual impairment who may use assistive technology to access the test content with speech software, screen reader or refreshable braille device, I want the content to be presented as intended, with accurate pronunciation and articulation, so that my assessment accurately reflects my knowledge of the content. 24 | 25 | C. As a student/learner with auditory and cognitive processing issues, it is difficult to distinguish sounds, inflections, and variations in pronunciation as rendered through synthetic voice, such as text-to-speech or screen reader technologies. Consistent and accurate pronunciation whether human-provided, external, or embedded is needed to support working executive processing, auditory processing and memory that facilitates comprehension in literacy and numeracy for learning and for assessments. 26 | 27 | D. As an English Learner (EL) or a visually impaired early learner using speech synthesis for reading comprehension that includes decoding words from letters as part of the learning construct (intent of measurement), pronunciation accuracy is vital to successful comprehension, as it allows the learner to distinguish sounds at the sentence, word, syllable, and phoneme level. 28 | 29 | ## Digital Content Management for TTS 30 | The advent of graphical user interfaces (GUIs) for the management and editing of text content has given rise to content creators not requiring technical expertise beyond the ability to operate a text editing application such as Microsoft Word. The following scenario summarizes the general use, accompanied by a hypothetical application. 31 | 32 | A. As a content creator, I want to create content that can readily be delivered through assistive technology, can convey the correct meaning, and ensure that screen readers render the right pronunciation based on the surrounding context. 33 | 34 | B. As a content producer for a global commercial site that is inclusive, I need to be able to provide accessible culture-specific content for different geographic regions. 35 | 36 | ### Educational Assessment 37 | In the educational assessment field, providing accurate and concise pronunciation for students with auditory accommodations, such as text-to-speech (TTS) or students with screen readers, is vital for ensuring content validity and alignment with the intended construct, which objectively measures a test takers knowledge and skills. For test administrators/educators, pronunciations must be consistent across instruction and assessment in order to avoid test bias or impact effects for students. Some additional requirements for the test administrators, include, but are not limited to, such scenarios: 38 | 39 | A. As a test administrator, I want to ensure that students with the read-aloud accommodation, who are using assistive technology or speech synthesis as an alternative to a human reader, have the same speech quality (e.g., intonation, expression, pronunciation, and pace, etc.) as a spoken language. 40 | 41 | B. As a math educator, I want to ensure that speech accuracy with mathematical expressions, including numbers, fractions, and operations have accurate pronunciation for those who rely on TTS. Some mathematical expressions require special pronunciations to ensure accurate interpretation while maintaining test validity and construct. Specific examples include: 42 | 43 | + Mathematical formulas written in simple text with special formatting should convey the correct meaning of the expression to identify changes from normal text to super- or to sub-script text. For example, without the proper formatting, the equation: 44 |a3-b3=(a-b)(a2+ab+b2) may incorrectly render through some technologies and applications as a3-b3=(a-b)(a2+ab+b2).
45 | + Distinctions made in writing are often not made explicit in speech; For example, “fx” may be interpreted as fx, f(x), fx, F X, F X. The distinction depends on the context; requiring the author to provide consistent and accurate semantic markup.
46 | + For math equations with Greek letters, it is important that the speech synthesizer be able to distinguish the phonetic differences between them, whether in the natural language or phonetic equivalents. For example, ε (epsilon) υ (upsilon) φ (phi) χ (chi) ξ(xi)
47 |
48 | C. As a test administrator/educator, pronunciations must be consistent across instruction and assessment, in order to avoid test bias and pronunciation effects on performance for students with disabilities (SWD) in comparison to students without disabilities (SWOD). Examples include:
49 |
50 | + If a test question is measuring rhyming of words or sounds of words, the speech synthesis should not read aloud the words, but rather spell out the words in the answer options.
51 | + If a test question is measuring spelling and the student needs to consider spelling correctness/incorrectness, the speech synthesis should not read aloud the misspelt words, especially for words, such as:
52 |
53 | i. *Heteronyms/homographs*: same spelling, different pronunciation, different meanings, such as lead (to go in front of) or lead (a metal); wind (to follow a course that is not straight) or wind (a gust of air); bass (low, deep sound) or bass (a type of fish), etc.
54 |
55 | ii. *Homophone*: words that sound alike, such as, to/two/too; there/their/they're; pray/prey; etc.
56 |
57 | iii. *Homonyms*: multiple meaning words, such as scale (measure) or scale (climb, mount); fair (reasonable) or fair (carnival); suit (outfit) or suit (harmonize); etc.
58 |
59 | ### Academic and Linguistic Practitioners
60 | The extension of content management in TTS is one as a means of encoding and preserving spoken text for academic analyses; irrespective of discipline, subject domain, or research methodology.
61 |
62 | A. As a linguist, I want to represent all the pronunciation variations of a given word in any language, for future analyses.
63 |
64 | B. As a speech language pathologist or speech therapists, I want TTS functionality to include components of speech and language that include dialectal and individual differences in pronunciation; identify differences in intonation, syntax, and semantics, and; allow for enhanced comprehension, language processing and support phonological awareness.
65 |
66 | ## Software Application Development
67 | Technical standards for software development assist organizations and individuals to provide accessible experiences for users with disabilities. The final user scenarios in this document are considered from the perspective of those who design and develop software.
68 |
69 | A. As a Product Owner for a web content management system (CMS), I want the next software product release to have the capability of pronouncing speech "just like Alexa can".
70 |
71 | B. As a client-side user interface developer, I need a way to render text content, so it is spoken accurately with assistive technologies.
72 |
73 |
74 |
75 |
76 |
77 |
--------------------------------------------------------------------------------
/user-scenarios/index.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 | The objective of the Pronunciation Task Force is to develop normative specifications and best practices guidance collaborating with other W3C groups as appropriate, to provide for proper pronunciation in HTML content when using text to speech (TTS) synthesis. This document provides various user scenarios highlighting the need for standardization of pronunciation markup, to ensure that consistent and accurate representation of the content. The requirements that come from the user scenarios provide the basis for the technical requirements/specifications.
17 |As part of the Accessible Platform Architectures (APA) Working Group, the Pronunciation Task Force (PTF) is a collaboration of subject matter experts working to identify and specify the optimal approach which can deliver reliably accurate pronunciation across browser and operating environments. With the introduction of the Kurzweil reading aid in 1976, to the more sophisticated synthetic speech currently used to assist communication as reading aids for the visually impaired and those with reading disabilities, the technology has multiple applications in education, communication, entertainment, etc. From helping to teach spelling and pronunciation in different languages, Text-to-Speech (TTS) has become a vital technology for providing access to digital content on the web and through mobile devices. 26 |
27 |The challenges that TTS presents include but are not limited to: the inability to accommodate regional variation and presentation of every phoneme present throughout the world; the incorrect determination by TTS of the pronunciation of content in context, and; the current inability to influence other pronunciation characteristics such as prosody and emphasis.
28 | 29 | 30 | 31 |The purpose of developing user scenarios is to facilitate discussion and further requirements definition for pronunciation standards developed within the PTF prior to review of the APA. There are numerous interpretations of what form user scenarios adopt. Within the user experience research (UXR) body of practice, a user scenario is a written narrative related to the use of a service from the perspective of a user or user group. Importantly, the context of use is emphasized as is the desired outcome of use. There are potentially thousands of user scenarios for a technology such as TTS, however, the focus for the PTF is on the core scenarios that relate to the kinds of users who will engage with TTS.
36 |User scenarios, like Personas, represent a composite of real-world experiences. In the case of the PTF, the scenarios were derived from interviews of people who were end-consumers of TTS, as well as submitted narratives and industry examples from practitioners. There are several formats of scenarios. Several are general goal or task-oriented scenarios. Others elaborate on richer context, for example, educational assessment.
37 |The following user scenarios are organized on the three perspectives of TTS use derived from analysis of the qualitative data collected from the discovery work:
38 |Ultimately, the quality and variation of TTS rendering by assistive technologies vary widely according to a user's context. The following user scenarios reinforce the necessity for accurate pronunciation from the perspective of those who consume digitally generated content.
46 |The advent of graphical user interfaces (GUIs) for the management and editing of text content has given rise to content creators not requiring technical expertise beyond the ability to operate a text editing application such as Microsoft Word. The following scenario summarizes the general use, accompanied by a hypothetical application.
57 |In the educational assessment field, providing accurate and concise pronunciation for students with auditory accommodations, such as text-to-speech (TTS) or students with screen readers, is vital for ensuring content validity and alignment with the intended construct, which objectively measures a test takers knowledge and skills. For test administrators/educators, pronunciations must be consistent across instruction and assessment in order to avoid test bias or impact effects for students. Some additional requirements for the test administrators, include, but are not limited to, such scenarios:
65 | 66 |a3-b3=(a-b)(a2+ab+b2) may incorrectly render through some technologies and applications as a3-b3=(a-b)(a2+ab+b2).The extension of content management in TTS is one as a means of encoding and preserving spoken text for academic analyses; irrespective of discipline, subject domain, or research methodology.
96 |Technical standards for software development assist organizations and individuals to provide accessible experiences for users with disabilities. The final user scenarios in this document are considered from the perspective of those who design and develop software.
106 |You say, pecan. I say, pecan.
71 | ``` 72 | 73 | Client will parse XML and serialize it before passing to a speech API: 74 | 75 | ```js 76 | var msg = new SpeechSynthesisUtterance(); 77 | var xml = document.getElementById('pecan').content.firstElementChild; 78 | msg.text = serialize(xml); 79 | speechSynthesis.speak(msg); 80 | ``` 81 | 82 | - `aria-ssml` referencing an XML string as script tag 83 | 84 | ```html 85 | 96 | 97 |You say, pecan. I say, pecan.
98 | ``` 99 | 100 | Client will pass the XML string raw to a speech API. 101 | 102 | ```js 103 | var msg = new SpeechSynthesisUtterance(); 104 | msg.text = document.getElementById('pecan').textContent; 105 | speechSynthesis.speak(msg); 106 | ``` 107 | 108 | - `aria-ssml` referencing an external XML document by URL 109 | 110 | ```html 111 |You say, pecan. I say, pecan.
112 | ``` 113 | 114 | Client will pass the string payload to a speech API. 115 | 116 | ```js 117 | var msg = new SpeechSynthesisUtterance(); 118 | var response = await fetch(el.dataset.ssml) 119 | msg.txt = await response.text(); 120 | speechSynthesis.speak(msg); 121 | ``` 122 | 123 | ### Existing Work 124 | 125 | - [`aria-ssml` proposal](https://github.com/alia11y/SSMLinHTMLproposal) 126 | - [SSML](https://www.w3.org/TR/speech-synthesis11/) 127 | - [Web Speech API](https://w3c.github.io/speech-api/) 128 | 129 | ### Problems and Limitations 130 | 131 | - `aria-ssml` is not a valid `aria-*` attribute. 132 | - OS/Browsers combinations that do not support the serialized XML usage of the Web Speech API. 133 | 134 | ## Use Case `data-ssml` 135 | 136 | ### Name 137 | `data-ssml` 138 | 139 | ### Owner 140 | Paul Grenier 141 | 142 | ### Background and Current Practice 143 | As an existing attribute, [`data-*`](https://html.spec.whatwg.org/multipage/dom.html#embedding-custom-non-visible-data-with-the-data-*-attributes) could be used, with some conventions, to include pronunciation content. 144 | 145 | ### Goal 146 | 147 | - Support repeated use within the page context 148 | - Support external file references 149 | - Reuse existing techniques without expanding specifications 150 | 151 | ### Target Audience 152 | 153 | - Hearing users 154 | 155 | ### Implementation Options 156 | 157 | - `data-ssml` as embedded JSON 158 | 159 | When an element with `data-ssml` is encountered by an SSML-aware AT, the AT should enhance the user interface by processing the referenced SSML content and passing it to the [Web Speech API](https://w3c.github.io/speech-api/) or an external API (e.g., [Google's Text to Speech API](https://cloud.google.com/text-to-speech/)). 160 | 161 | 162 | ```html 163 | I say pecan. 164 | You say pecan. 165 | ``` 166 | 167 | Client will convert JSON to SSML and pass the XML string a speech API. 168 | 169 | ```js 170 | var msg = new SpeechSynthesisUtterance(); 171 | msg.text = convertJSONtoSSML(element.dataset.ssml); 172 | speechSynthesis.speak(msg); 173 | ``` 174 | 175 | - `data-ssml` referencing XML by template ID 176 | 177 | ```html 178 | 179 | 180 | 181 |You say, pecan. I say, pecan.
193 | ``` 194 | 195 | Client will parse XML and serialize it before passing to a speech API: 196 | 197 | ```js 198 | var msg = new SpeechSynthesisUtterance(); 199 | var xml = document.getElementById('pecan').content.firstElementChild; 200 | msg.text = serialize(xml); 201 | speechSynthesis.speak(msg); 202 | ``` 203 | 204 | - `data-ssml` referencing an XML string as script tag 205 | 206 | ```html 207 | 218 | 219 |You say, pecan. I say, pecan.
220 | ``` 221 | 222 | Client will pass the XML string raw to a speech API. 223 | 224 | ```js 225 | var msg = new SpeechSynthesisUtterance(); 226 | msg.text = document.getElementById('pecan').textContent; 227 | speechSynthesis.speak(msg); 228 | ``` 229 | 230 | - `data-ssml` referencing an external XML document by URL 231 | 232 | ```html 233 |You say, pecan. I say, pecan.
234 | ``` 235 | 236 | Client will pass the string payload to a speech API. 237 | 238 | ```js 239 | var msg = new SpeechSynthesisUtterance(); 240 | var response = await fetch(el.dataset.ssml) 241 | msg.txt = await response.text(); 242 | speechSynthesis.speak(msg); 243 | ``` 244 | 245 | ### Existing Work 246 | 247 | - [`aria-ssml` proposal](https://github.com/alia11y/SSMLinHTMLproposal) 248 | - [SSML](https://www.w3.org/TR/speech-synthesis11/) 249 | - [Web Speech API](https://w3c.github.io/speech-api/) 250 | 251 | ### Problems and Limitations 252 | 253 | - Does not assume or suggest visual pronunciation help for deaf or hard of hearing 254 | - Use of `data-*` requires input from AT vendors 255 | - XML data is not indexed by search engines 256 | 257 | ## Use Case HTML5 258 | 259 | ### Name 260 | HTML5 261 | 262 | ### Owner 263 | Paul Grenier 264 | 265 | ### Background and Current Practice 266 | HTML5 includes the XML [namespaces](https://www.w3.org/TR/html5/infrastructure.html#namespaces) for MathML and SVG. So, using either's elements in an HTML5 document is valid. Because SSML's implementation is non-visual in nature, browser implementation could be slow or non-existent without affecting how authors use SSML in HTML. Expansion of HTML5 to include SSML namespace would allow valid use of SSML in the HTML5 document. Browsers would treat the element like any other unknown element, as [`HTMLUnknownElement`](https://www.w3.org/TR/html50/dom.html#htmlunknownelement). 267 | 268 | ### Goal 269 | 270 | - Support valid use of SSML in HTML5 documents 271 | - Allow visual pronunciation support 272 | 273 | ### Target Audience 274 | 275 | - SSML-aware technologies and browser extensions 276 | - Search indexers 277 | 278 | ### Implementation Options 279 | 280 | - SSML 281 | 282 | When an element with [`data-ssml`](https://www.w3.org/TR/wai-aria-1.1/#aria-details) is encountered by an [SSML](https://www.w3.org/TR/speech-synthesis11/)-aware AT, the AT should enhance the user interface by processing the referenced SSML content and passing it to the [Web Speech API](https://w3c.github.io/speech-api/) or an external API (e.g., [Google's Text to Speech API](https://cloud.google.com/text-to-speech/)). 283 | 284 | ```html 285 |` tag from SSML is not given the prefix because we still want to start a semantic paragraph within the content. The other tags used in the example have no semantic meaning. Tags like `` in HTML could be converted to `
340 |
354 | Speak
355 |
359 |
444 | Do you listen to WKRP?
447 |
491 | You say,
492 |
493 |
494 | pecan
495 |
496 |
497 | .
498 |
499 | I say,
500 |
501 |
502 | pe
503 |
504 | can
505 |
506 |
507 | .
508 |
509 | This document is the Gap Analysis Review which presents required features of Spoken Text Pronunciation and
14 | Presentation and existing standards or specifications that may support (or enable support) of those features. Gaps
15 | are defined when a
16 | required feature does not have a corresponding method by which it can be authored in HTML. Accurate, consistent pronunciation and presentation of content spoken by text to speech synthesis (TTS) is an
22 | essential requirement in education and other domains. Organizations such as educational publishers and assessment
23 | vendors are
24 | looking for a standards-based solution to enable authoring of spoken presentation guidance in HTML which can then
25 | be consumed by assistive technologies and other applications which utilize text to speech synthesis for rendering
26 | of content. W3C has developed two standards pertaining to the presentation of speech synthesis which have reached
28 | recommendation status, Speech Synthesis Markup Language
29 | (SSML) and the Pronunciation Lexicon Specification
30 | (PLS). Both standards are directly consumed by a speech synthesis engine supporting those standards. While a PLS
31 | file reference may be referenced in a HTML
32 | page using CSS Speech Module is a retired W3C Working Group Note that
38 | describes mechanism by which content authors may apply a variety of speech styling and presentation properties to
39 | HTML. This approach
40 | has a variety of advantages but does not implement the full set of features required for pronunciation. Section 16 of the Note specifically references the
42 | issue of pronunciation: While a portion of CSS Speech was demonstrated by Apple in 2011 on iOS with Safari and VoiceOver,
55 | it is not presently supported on any platform with any Assistive Technology,
56 | and work on the standard has itself been stopped by the CSS working group. Efforts to address this need have been considered by both assessment technology vendors and the publishing
58 | community. Citing the need for pronunciation and presentation controls, the IMS Global Learning Consortium added
59 | the ability to author SSML
60 | markup, specify PLS files, and reference CSS Speech properties to the Question Test Interoperability (QTI) Accessible Portable Item Protocol (APIP).
62 | In practice, QTI/APIP
63 | authored content is transformed into HTML for rendering in web browsers. This led to the dilemma that there is no
64 | standardized (and supported) method for inlining SSML in HTML, nor is there support for CSS Speech. This has led
65 | to the situation
66 | where SSML is the primary authoring model, with assessment vendors implementing a custom method for adding the
67 | SSML (or SSML-like) features to HTML using non-standard or data attributes, with customized Read Aloud software
68 | consuming those
69 | attributes for text to speech synthesis. Given the need to deliver accurate spoken presentation, non-standard
70 | approaches often include mis-use of WAI-ARIA, and novel or contextually non-valid attributes (e.g.,
71 | The attribute model for adding pronunciation and presentation guidance for assistive technologies and text to
78 | speech synthesis has demonstrated traction by vendors trying to solve this need. It should be noted that many of
79 | the required
80 | features are not well supported by a single attribute, as most follow the form of a presentation property / value
81 | pairing. Using multiple attributes to provide guidance to assistive technologies is not novel, as seen with
82 | WAI-ARIA where multiple
83 | attributes may be applied to a single element, for example, The common spoken pronunciation requirements from the education domain serve as a primary source for these
95 | features. These requirements can be broken down into the following main functions that would support authoring and
96 | spoken presentation
97 | needs. When content is authored in mixed language, a mechanism is needed to allow authors to indicate both the base
101 | language of the content as well as the language of individual words and phrases. The expectation is that
102 | assistive technologies and
103 | other tools that utilize text to speech synthesis would detect and apply the language requested when presenting
104 | the text. Content authors may elect to make adjustments of those paramters to control the spoken presentation for
109 | purposes such as providing a gender specific voice to reflect that of the author, or for a character (or
110 | characters) in theatrical
111 | presentation of a story. Many assistive technologies already provide user selection of voice family and gender
112 | independent of any authored intent. In some cases words may need to have their phonetic pronunciation prescribed by the content author. This may
117 | occur when uncommon words (not supported by text to speech synthesizers), or in cases where word pronunciation
118 | will vary based on
119 | context, and that context may not be correctly described. There are cases where content that is visually presented may require replacement (substitution) with an
124 | alternate textual form to ensure correct pronunciation by text to speech synthesizers. In some cases phonetic
125 | pronunciation may be a
126 | solution to this need. While end users should have full control over spoken presentation parameters such as speaking rate, pitch, and
131 | volume (e.g., WCAG 1.4.2 ), content authors may elect to make adjustments of those parameters to control the
132 | spoken presentation for
133 | purposes such as a theatrical presentation of a story. Many assistive technologies already provide user control
134 | speaking rate, pitch, and volume independent of any authored intent. In written text, an author may find it necessary to add emphasis to an important word or phrase. HTML supports
139 | both semantic elements (e.g., While text to speech engines continue to improve in their ability to process text and provide accurate spoken
149 | rendering of acronyms and numeric values, there can be instances where uncommon terms or alphanumeric constructs
150 | pose challenges.
151 | Further, some educators may have specific requirements as to how a numeric value be spoken which may differ from
152 | a TTS engine's default rendering. For example, the Smarter Balanced Assessment Consortium has developed Read Aloud Guidelines to be
154 | followed by human readers used by students who may require a spoken presentation of an educational test, which
155 | includes specific examples
156 | of how numeric values should be read aloud. Precise control as to how numeric values should be spoken may not always be correctly determined by text to
160 | speech engines from context. Examples include speaking a number as individual digits, correct reading of
161 | year values, and
162 | the correct speaking of ordinal and cardinal numbers. Precise control as to how string values should be spoken, which may not be determined correctly by text to
167 | speech synthesizers. Specific spoken presentation requirements exist in the Accessibility Guidelines from PARCC, and
173 | include requirements such as inserting pauses in the spoken presentation, before and after emphasized words and
174 | mathematical
175 | terms. In practice, content authors may find it necessary to insert pauses between numeric values to limit the
176 | chance of hearing multiple numbers as a single value. One common technique to achieve pausing to date has
177 | involved inserting
178 | non-visible commas before or after a text string requiring a pause. While this may work in practice for a read
179 | aloud TTS tool, it is problematic for screen reader users who may, based on verbosity settings, hear the
180 | multiple commas announced,
181 | and for refreshable braille users who will have the commas visible in braille. Based on the features and use cases described in the prior sections, the following table presents existing speech
188 | presentation standards, HTML features, and WAI-ARIA attributes that may offer a method to achieve the requirement
189 | for HTML authors. A blank cell for any approach represents a gap in support. The following sections describe how each of the required features may be met by the use of existing approaches. A
267 | key consideration in the analysis is whether a means exists to directly author (or annotate) HTML content to
268 | incorporate the
269 | spoken presentation and pronunciation feature. Allow content authors to specify the language of text contained within an element so that the TTS used for
273 | rendering will select the appropriate language for synthesis. Example: Allow content authors to specify a specific TTS voice to be used to render text. For example, for content that
285 | presents a dialog between two people, a woman and a man, the author may specify that a female voice be used for
286 | the woman's text and a
287 | male voice be used for the man's text. Some platform TTS services may support a variety of voices, identified by
288 | a name, gender, or even age. Example: Using the Example: Allow content authors to precisely specify the phonetic pronunciation of a word or phrase. Using PLS, all the pronunciations can be factored out into an external PLS document which is referenced by the
303 |
306 | Introduction
21 | link rel, there is no known uptake of PLS using this method by assistive technologies.
33 | While there are technically methods to allow authors to inline SSML within HTML (using namespaces), such an
34 | approach has not been
35 | adopted, and anecdotal comments from browser and assistive technology vendors have suggested this is not a viable
36 | approach.
44 | CSS does not specify how to define the pronunciation (expressed using a well-defined phonetic alphabet) of a
45 | particular piece of text within the markup document. A "phonemes" property was described in earlier drafts of this
46 | specification, but
47 | objections were raised due to breaking the principle of separation between content and presentation (the
48 | "phonemes" authored within aural CSS stylesheets would have needed to be updated each time text changed within the
49 | markup document). The
50 | "phonemes" functionality is therefore considered out-of-scope in CSS (the presentation layer) and should be
51 | addressed in the markup / content layer.
52 |
53 | label). A
72 | particular problem occurs when custom pronunciation is applied via a misuse of the aria-label
73 | attribute, which results in an issue for screen reader users who also rely upon refreshable braille, and in which
74 | a hinted pronunciation
75 | intended only for a text to speech synthesizer also appears on the braille display.
76 | role and aria-checked. The
84 | EPUB
85 | standard for
86 | digital publishing introduced a namespaced version of the SSML phoneme and alphabet
87 | attributes enabling content authors to provide pronunciation guidance. Uptake by the publishing community has been
88 | limited, reportedly
89 | due to the lack of support in reading systems and assistive technologies.
90 | Core Features for Pronunciation and Spoken Presentation
94 | Language
100 | Voice Family / Gender
108 | Phonetic Pronunciation of String Values
116 | String Substitution
123 | Rate / Pitch / Volume
130 | Emphasis
138 | em) and CSS properties which, through a variety of style options,
140 | make programmatic
141 | detection of authored emphasis difficult (e.g., font-weight: heavy). While the emphasis element has
142 | existed since HTML 2.0, there is currently no uptake by assistive technology or read aloud tools to present text
143 | semantically tagged
144 | for emphasis to be spoken with emphasis.Say As
148 | Presentation of Numeric Values
159 | Presentation of String Values
166 | Pausing
172 | Gap Analysis
187 |
191 |
192 |
266 |
193 |
200 | Requirement
194 | HTML
195 | WAI-ARIA
196 | PLS
197 | CSS Speech
198 | SSML
199 |
201 |
208 | Language
202 | Yes
203 | Yes
207 |
209 |
216 | Voice Family/Gender
210 | Yes
214 | Yes
215 |
217 |
224 | Phonetic Pronunciation
218 | Yes
221 | Yes
223 |
225 |
232 | Substitution
226 | Partial
228 | Yes
231 |
233 |
240 | Rate/Pitch/Volume
234 | Yes
238 | Yes
239 |
241 |
248 | Emphasis
242 | Yes
243 | Yes
246 | Yes
247 |
249 |
256 | Say As
250 | Yes
255 |
257 |
264 |
265 | Pausing
258 | Yes
262 | Yes
263 | Language
272 | HTML
276 | lang attribute can be applied at the document level or to individual elements. (WCAG) (AT
277 | Supported: some)SSML
279 | <speak> In Paris, they pronounce it <lang xml:lang="fr-FR">Paris</lang>
280 | </speak>code>Voice Family/Gender
284 | CSS
290 | voice-family property can be used to specify the gender of the voice.{ voice-family: male; }SSML
293 | <voice> element, the gender of the speaker, if supported by the TTS engine, can be
294 | specified.<voice gender="female" >Mary had a little lamb,</voice>Phonetic Pronunciation
300 | PLS
302 | <lexicon> element of SSML
304 | Example:
310 | <speak> <lexicon uri="http://www.example.com/movie_lexicon.pls"/>
307 | The title of the movie is: "La vita è bella" (Life is beautiful),
308 | which is directed by Roberto Benigni.</speak>
309 |
The following is a simple example of an SSML document. It includes an Italian movie title and the name of the 314 | director to be read in US English.
315 |316 |
Example: The title of the movie is:
317 | <speak> <phoneme alphabet="ipa" ph="ˈlɑ ˈviːɾə ˈʔeɪ ˈbɛlə">
318 | "La vita è bella"</phoneme> (Life is beautiful),
319 | which is directed by
320 | <phoneme alphabet="ipa" ph="ɹəˈbɛːɹɾoʊ bɛˈniːnji"">
321 | Roberto Benigni </phoneme>.</speak>
322 |
323 |
324 |
325 |
326 |
327 | Allow content authors to substitute a text string to be rendered by TTS instead of the actual text contained in 330 | an element.
331 |The aria-label and aria-labelledby attribute can be used
334 | by an author to supply a text string
335 | that will become the accessible name for the element upon which it is applied. This usage effectively
336 | provides a mechanism for performing text substation that is supported by a screen reader. However, it is
337 | problematic for one significant reason; for users who utilize screen readers and refreshable Braille, the
338 | content that is voiced will not match the content that is sent to the refreshable Braille device. This mismatch
339 | would not be acceptable for some content, particularly for assessment content.
Pronounce the specified word or phrase as a different word or phrase. Specify the pronunciation to substitute
342 | with the alias attribute.
344 |
345 |
346 | <speak>
347 | My favorite chemical element is <sub alias="aluminum">Al</sub>,
348 | but Al prefers <sub alias="magnesium">Mg</sub>.
349 | </speak>
350 |
351 |
352 |
353 | Allow content authors to specify characteristics, such as rate, pitch, and/or volume of the TTS rendering of 357 | the text.
358 |360 |
‘voice-rate’ property manipulates the rate of generated synthetic speech in terms of words
363 | per minute.368 |
‘voice-pitch’ property specifies the "baseline" pitch of the generated speech output, which
371 | depends on the used ‘voice-family’ instance, and varies across speech synthesis processors (it
372 | approximately corresponds to the average pitch of the output). For example, the common pitch for a male voice
373 | is around 120Hz, whereas it is around 210Hz for a female voice.377 |
‘voice-range’ property specifies the variability in the "baseline" pitch, i.e. how much the
380 | fundamental frequency may deviate from the average pitch of the speech output. The dynamic pitch range of the
381 | generated speech generally increases for a highly animated voice, for example when variations in inflection
382 | are used to convey meaning and emphasis in speech. Typically, a low range produces a flat, monotonic voice,
383 | whereas a high range produces an animated voice.prosody modifies the volume, pitch, and rate of the tagged speech.
389 |
390 |
391 | <speak>
392 | Normal volume for the first sentence.
393 | <prosody volume="x-loud">Louder volume for the second sentence</prosody>.
394 | When I wake up, <prosody rate="x-slow">I speak quite slowly</prosody>.
395 | I can speak with my normal pitch,
396 | <prosody pitch="x-high"> but also with a much higher pitch </prosody>,
397 | and also <prosody pitch="low">with a lower pitch</prosody>.
398 | </speak>
399 |
400 |
401 |
402 | Allow content authors to specify that text content be spoken with emphasis, for example, louder and more 406 | slowly. This can be viewed as a simplification of the Rate/Pitch/Volume controls to reduce authoring complexity. 407 |
408 |
410 | The HTML <em> element marks text that has stress emphasis. The <em>
411 | element can be nested, with each level of nesting indicating a greater degree of emphasis.
412 |
414 | The <em> element is for words that have a stressed emphasis compared to surrounding text,
415 | which is often limited to a word or words of a sentence and affects the meaning of the sentence itself.
416 |
417 | Typically this element is displayed in italic type. However, it should not be used simply to apply italic
418 | styling; use the CSS font-style property for that purpose. Use the <cite>
419 | element to mark the title of a work (book, play, song, etc.). Use the <i> element to mark
420 | text that is in an alternate tone or mood, which covers many common situations for italics such as scientific
421 | names or words in other languages. Use the <strong> element to mark text that has greater
422 | importance than surrounding text.
423 |
427 |
‘voice-stress’ property manipulates the strength of emphasis, which is normally applied
430 | using a combination of pitch change, timing changes, loudness and other acoustic differences. The precise
431 | meaning of the values therefore depend on the language being spoken.Emphasize the tagged words or phrases. Emphasis changes rate and volume of the speech. More emphasis is spoken 437 | louder and slower. Less emphasis is quieter and faster.
438 |439 |
440 |
441 | <speak>
442 | I already told you I
443 | <emphasis level="strong">really like</emphasis> that person.
444 | </speak>
445 |
446 |
447 |
448 | Allow content authors to specify how text is spoken. For example, content authors would be able to indicate 452 | that a series of four numbers should be spoken as a year rather than a cardinal number.
453 |The ‘speak-as’ property determines in what manner text gets rendered aurally, based upon a
455 | predefined list of possibilities.
457 | Speech synthesizers are knowledgeable about what a number is. The ‘speak-as’ property enables some
458 | level of control on how user agents render numbers, and may be implemented as a preprocessing step before
459 | passing the text to the actual speech synthesizer.
460 |
463 | Describes how the text should be interpreted. This lets you provide additional context to the text and eliminate
464 | any ambiguity on how Alexa should render the text. Indicate how Alexa should interpret the text with the
465 | interpret-as attribute.
466 |
468 |
469 |
470 | <speak>
471 | Here is a number spoken as a cardinal number:
472 | <say-as interpret-as="cardinal">12345</say-as>.
473 | Here is the same number with each digit spoken separately:
474 | <say-as interpret-as="digits">12345</say-as>.
475 | Here is a word spelled out: <say-as interpret-as="spell-out">hello</say-as>
476 | </speak>
477 |
478 |
479 |
480 |
481 | Allow content authors to specify pauses before or after content to ensure the desired prosody of the 485 | presentation, which can affect the pronunciation of the pronunciation of content the precedes or follows the 486 | pause.
487 |
489 | The ‘pause-before’ and ‘pause-after’ properties specify a prosodic boundary (silence
490 | with a specific duration) that occurs before (or after) the speech synthesis rendition of the selected element,
491 | or if any ‘cue-before’ (or ‘cue-after’) is specified, before (or after) the cue within
492 | the aural box model.
493 |
495 |
496 | Note that although the functionality provided by this property is similar to the break element from
497 | the SSML markup language [SSML], the application of ‘pause’ prosodic boundaries within the aural box model of
498 | CSS Speech requires special considerations (e.g. "collapsed" pauses).
499 |
500 |
503 | break represents a pause in the speech. Set the length of the pause with the strength
504 | or time attributes.
505 |
507 |
508 |
509 | <speak>
510 | There is a three second pause here <break time="3s"/>
511 | then the speech continues.
512 | </speak>
513 |
514 |
515 |
516 |
517 | The objective of the Pronunciation Task Force is to develop normative specifications and best practices guidance collaborating with other W3C groups as appropriate, to provide for proper pronunciation in HTML content when using text to speech (TTS) synthesis. This document provides various use cases highlighting the need for standardization of pronunciation markup, to ensure that consistent and accurate representation of the content. The requirements from the user scenarios provide the basis for these technical requirements/specifications.
17 |This document provides use cases which describe specific implmentation approaches for introducing pronunciation 26 | and spoken presentation authoring markup into HTML5. These approaches are based on the two primary approaches 27 | that have evolved from the Pronunciation Task Force members. Other approaches may appear in subsequent working drafts. 28 |
29 |Successful use cases will be those that provide ease of authoring and consumption by assistive technologies and user 30 | agents that utilize synthetic speech for spoken presentation of web content. The most challenging aspect of consumption may 31 | be alignment of the markup approach with the standard mechanisms by which assistive technologies, specifically screen 32 | readers, obtain content via platform accessibility APIs. 33 |
34 | 35 | 36 |A new aria attribute could be used to include pronunciation content.
Embed SSML in an HTML document.
46 |aria-ssml as embedded JSON
58 |When AT encounters an element with aria-ssml, the AT should enhance the UI by processing the pronunciation content and passing it to the Web Speech API or an external API (e.g., Google's Text to Speech API).
59 |I say <span aria-ssml='{"phoneme":{"ph":"pɪˈkɑːn","alphabet":"ipa"}}'>pecan</span>.
60 | You say <span aria-ssml='{"phoneme":{"ph":"ˈpi.kæn","alphabet":"ipa"}}'>pecan</span>.
61 | Client will convert JSON to SSML and pass the XML string a speech API.
62 |var msg = new SpeechSynthesisUtterance();
63 | msg.text = convertJSONtoSSML(element.getAttribute('aria-ssml'));
64 | speechSynthesis.speak(msg);
65 | aria-ssml referencing XML by template ID
66 |<!-- ssml must appear inside a template to be valid --> 67 | <template id="pecan"> 68 | <?xml version="1.0"?> 69 | <speak version="1.1" 70 | xmlns="http://www.w3.org/2001/10/synthesis" 71 | xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" 72 | xsi:schemaLocation="http://www.w3.org/2001/10/synthesis 73 | http://www.w3.org/TR/speech-synthesis11/synthesis.xsd" 74 | xml:lang="en-US"> 75 | You say, <phoneme alphabet="ipa" ph="pɪˈkɑːn">pecan</phoneme>. 76 | I say, <phoneme alphabet="ipa" ph="ˈpi.kæn">pecan</phoneme>. 77 | </speak> 78 | </template> 79 | 80 | <p aria-ssml="#pecan">You say, pecan. I say, pecan.</p>81 |
Client will parse XML and serialize it before passing to a speech API:
82 |var msg = new SpeechSynthesisUtterance();
83 | var xml = document.getElementById('pecan').content.firstElementChild;
84 | msg.text = serialize(xml);
85 | speechSynthesis.speak(msg);
86 |
87 | aria-ssml referencing an XML string as script tag
88 |<script id="pecan" type="application/ssml+xml"> 89 | <speak version="1.1" 90 | xmlns="http://www.w3.org/2001/10/synthesis" 91 | xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" 92 | xsi:schemaLocation="http://www.w3.org/2001/10/synthesis 93 | http://www.w3.org/TR/speech-synthesis11/synthesis.xsd" 94 | xml:lang="en-US"> 95 | You say, <phoneme alphabet="ipa" ph="pɪˈkɑːn">pecan</phoneme>. 96 | I say, <phoneme alphabet="ipa" ph="ˈpi.kæn">pecan</phoneme>. 97 | </speak> 98 | </script> 99 | 100 | <p aria-ssml="#pecan">You say, pecan. I say, pecan.</p>101 |
Client will pass the XML string raw to a speech API.
102 |var msg = new SpeechSynthesisUtterance();
103 | msg.text = document.getElementById('pecan').textContent;
104 | speechSynthesis.speak(msg);
105 | aria-ssml referencing an external XML document by URL
106 |<p aria-ssml="http://example.com/pronounce.ssml#pecan">You say, pecan. I say, pecan.</p>107 |
Client will pass the string payload to a speech API.
108 |var msg = new SpeechSynthesisUtterance(); 109 | var response = await fetch(el.dataset.ssml) 110 | msg.txt = await response.text(); 111 | speechSynthesis.speak(msg);112 |
As an existing attribute, data-* could be used, with some conventions, to include pronunciation content.
136 |Hearing users
148 |data-ssml as embedded JSON
152 |When an element with data-ssml is encountered by an SSML-aware AT, the AT should enhance the user interface by processing the referenced SSML content and passing it to the Web Speech API or an external API (e.g., Google's Text to Speech API).
153 |<h2>The Pronunciation of Pecan</h2>
154 | <p><speak>
155 | I say <span data-ssml='{"phoneme":{"ph":"pɪˈkɑːn","alphabet":"ipa"}}'>pecan</span>.
156 | You say <span data-ssml='{"phoneme":{"ph":"ˈpi.kæn","alphabet":"ipa"}}'>pecan</span>.
157 | Client will convert JSON to SSML and pass the XML string a speech API.
158 |var msg = new SpeechSynthesisUtterance(); 159 | msg.text = convertJSONtoSSML(element.dataset.ssml); 160 | speechSynthesis.speak(msg);161 | 162 |
data-ssml referencing XML by template ID
163 |<!-- ssml must appear inside a template to be valid --> 164 | <template id="pecan"> 165 | <?xml version="1.0"?> 166 | <speak version="1.1" 167 | xmlns="http://www.w3.org/2001/10/synthesis" 168 | xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" 169 | xsi:schemaLocation="http://www.w3.org/2001/10/synthesis 170 | http://www.w3.org/TR/speech-synthesis11/synthesis.xsd" 171 | xml:lang="en-US"> 172 | You say, <phoneme alphabet="ipa" ph="pɪˈkɑːn">pecan</phoneme>. 173 | I say, <phoneme alphabet="ipa" ph="ˈpi.kæn">pecan</phoneme>. 174 | </speak> 175 | </template> 176 | 177 | <p data-ssml="#pecan">You say, pecan. I say, pecan.</p>178 |
Client will parse XML and serialize it before passing to a speech API:
179 |var msg = new SpeechSynthesisUtterance();
180 | var xml = document.getElementById('pecan').content.firstElementChild;
181 | msg.text = serialize(xml);
182 | speechSynthesis.speak(msg);
183 |
184 | data-ssml referencing an XML string as script tag
185 |<script id="pecan" type="application/ssml+xml"> 186 | <speak version="1.1" 187 | xmlns="http://www.w3.org/2001/10/synthesis" 188 | xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" 189 | xsi:schemaLocation="http://www.w3.org/2001/10/synthesis 190 | http://www.w3.org/TR/speech-synthesis11/synthesis.xsd" 191 | xml:lang="en-US"> 192 | You say, <phoneme alphabet="ipa" ph="pɪˈkɑːn">pecan</phoneme>. 193 | I say, <phoneme alphabet="ipa" ph="ˈpi.kæn">pecan</phoneme>. 194 | </speak> 195 | </script> 196 | 197 | <p data-ssml="#pecan">You say, pecan. I say, pecan.</p>198 |
Client will pass the XML string raw to a speech API.
199 |var msg = new SpeechSynthesisUtterance();
200 | msg.text = document.getElementById('pecan').textContent;
201 | speechSynthesis.speak(msg);
202 | data-ssml referencing an external XML document by URL
203 |<p data-ssml="http://example.com/pronounce.ssml#pecan">You say, pecan. I say, pecan.</p>204 |
Client will pass the string payload to a speech API.
205 |var msg = new SpeechSynthesisUtterance(); 206 | var response = await fetch(el.dataset.ssml) 207 | msg.txt = await response.text(); 208 | speechSynthesis.speak(msg);209 |
HTML5 includes the XML namespaces for MathML and SVG. So, using either's elements in an HTML5 document is valid. Because SSML's implementation is non-visual in nature, browser implementation could be slow or non-existent without affecting how authors use SSML in HTML. Expansion of HTML5 to include SSML namespace would allow valid use of SSML in the HTML5 document. Browsers would treat the element like any other unknown element, as HTMLUnknownElement.
233 |SSML
251 |When an element with data-ssml is encountered by an SSML-aware AT, the AT should enhance the user interface by processing the referenced SSML content and passing it to the Web Speech API or an external API (e.g., Google's Text to Speech API).
252 |<h2>The Pronunciation of Pecan</h2> 253 | <p><speak> 254 | You say, <phoneme alphabet="ipa" ph="pɪˈkɑːn">pecan</phoneme>. 255 | I say, <phoneme alphabet="ipa" ph="ˈpi.kæn">pecan</phoneme>. 256 | </speak></p>257 |
SSML is not valid HTML5
269 |Embed valid SSML in HTML using custom elements registered as ssml-* where * is the actual SSML tag name (except for p which expects the same treatment as an HTML p in HTML layout).
277 |Support use of SSML in HTML documents.
281 |ssml-speak: see demo
292 |Only the <ssml-speak> component requires registration. The component code lifts the SSML by getting the innerHTML and removing the ssml- prefix from the interior tags and passing it to the web speech API. The <p> tag from SSML is not given the prefix because we still want to start a semantic paragraph within the content. The other tags used in the example have no semantic meaning. Tags like <em> in HTML could be converted to <emphasis> in SSML. In that case, CSS styles will come from the browser's default styles or the page author.
293 |<ssml-speak>
294 | Here are <ssml-say-as interpret-as="characters">SSML</ssml-say-as> samples.
295 | I can pause<ssml-break time="3s"></ssml-break>.
296 | I can speak in cardinals.
297 | Your number is <ssml-say-as interpret-as="cardinal">10</ssml-say-as>.
298 | Or I can speak in ordinals.
299 | You are <ssml-say-as interpret-as="ordinal">10</ssml-say-as> in line.
300 | Or I can even speak in digits.
301 | The digits for ten are <ssml-say-as interpret-as="characters">10</ssml-say-as>.
302 | I can also substitute phrases, like the <ssml-sub alias="World Wide Web Consortium">W3C</ssml-sub>.
303 | Finally, I can speak a paragraph with two sentences.
304 | <p>
305 | <ssml-s>You say, <ssml-phoneme alphabet="ipa" ph="pɪˈkɑːn">pecan</ssml-phoneme>.</ssml-s>
306 | <ssml-s>I say, <ssml-phoneme alphabet="ipa" ph="ˈpi.kæn">pecan</ssml-phoneme>.</ssml-s>
307 | </p>
308 | </ssml-speak>
309 | <template id="ssml-controls">
310 | <style>
311 | [role="switch"][aria-checked="true"] :first-child,
312 | [role="switch"][aria-checked="false"] :last-child {
313 | background: #000;
314 | color: #fff;
315 | }
316 | </style>
317 | <slot></slot>
318 | <p>
319 | <span id="play">Speak</span>
320 | <button role="switch" aria-checked="false" aria-labelledby="play">
321 | <span>on</span>
322 | <span>off</span>
323 | </button>
324 | </p>
325 | </template>
326 | class SSMLSpeak extends HTMLElement {
327 | constructor() {
328 | super();
329 | const template = document.getElementById('ssml-controls');
330 | const templateContent = template.content;
331 | this.attachShadow({mode: 'open'})
332 | .appendChild(templateContent.cloneNode(true));
333 | }
334 | connectedCallback() {
335 | const button = this.shadowRoot.querySelector('[role="switch"][aria-labelledby="play"]')
336 | const ssml = this.innerHTML.replace(/ssml-/gm, '')
337 | const msg = new SpeechSynthesisUtterance();
338 | msg.lang = document.documentElement.lang;
339 | msg.text = `<speak version="1.1"
340 | xmlns="http://www.w3.org/2001/10/synthesis"
341 | xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
342 | xsi:schemaLocation="http://www.w3.org/2001/10/synthesis
343 | http://www.w3.org/TR/speech-synthesis11/synthesis.xsd"
344 | xml:lang="${msg.lang}">
345 | ${ssml}
346 | </speak>`;
347 | msg.voice = speechSynthesis.getVoices().find(voice => voice.lang.startsWith(msg.lang));
348 | msg.onstart = () => button.setAttribute('aria-checked', 'true');
349 | msg.onend = () => button.setAttribute('aria-checked', 'false');
350 | button.addEventListener('click', () => speechSynthesis[speechSynthesis.speaking ? 'cancel' : 'speak'](msg))
351 | }
352 | }
353 |
354 | customElements.define('ssml-speak', SSMLSpeak);
355 | JSON-LD provides an established standard for embedding data in HTML. Unlike other microdata approaches, JSON-LD helps to reuse standardized annotations through external references.
379 |Support use of SSML in HTML documents.
383 |JSON-LD
394 |<script type="application/ld+json">
395 | {
396 | "@context": "http://schema.org/",
397 | "@id": "/pronunciation#WKRP",
398 | "@type": "RadioStation",
399 | "name": ["WKRP",
400 | "@type": "PronounceableText",
401 | "textValue": "WKRP",
402 | "speechToTextMarkup": "SSML",
403 | "phoneticText": "<speak><say-as interpret-as=\"characters\">WKRP</say-as>"
404 | ]
405 | }
406 | </script>
407 | <p>
408 | Do you listen to <span itemscope
409 | itemtype="http://schema.org/PronounceableText"
410 | itemid="/pronunciation#WKRP">WKRP</span>?
411 | </p>
412 | not an established "type"/published schema
427 |<Ruby> annotations are short runs of text presented alongside base text, primarily used in East Asian typography as a guide for pronunciation or to include other annotations.434 |
ruby guides pronunciation visually. This seems like a natural fit for text-to-speech.
435 |ruby with microdata
453 |Microdata can augment the ruby element and its descendants.
454 |<p> 455 | You say, 456 | <span itemscope="" itemtype="http://example.org/Pronunciation"> 457 | <ruby itemprop="phoneme" content="pecan"> 458 | pecan 459 | <rt itemprop="ph">pɪˈkɑːn</rt> 460 | <meta itemprop="alphabet" content="ipa"> 461 | </ruby>. 462 | </span> 463 | I say, 464 | <span itemscope="" itemtype="http://example.org/Pronunciation"> 465 | <ruby itemprop="phoneme" content="pecan"> 466 | pe 467 | <rt itemprop="ph">ˈpi</rt> 468 | can 469 | <rt itemprop="ph">kæn</rt> 470 | <meta itemprop="alphabet" content="ipa"> 471 | </ruby>. 472 | </span> 473 | </p>474 |
The purpose of developing user scenarios is to facilitate discussion and further requirements definition for pronunciation standards developed within the PTF prior to review of the APA. There are numerous interpretations of what form user scenarios adopt. Within the user experience research (UXR) body of practice, a user scenario is a written narrative related to the use of a service from the perspective of a user or user group. Importantly, the context of use is emphasized as is the desired outcome of use. There are potentially thousands of user scenarios for a technology such as TTS, however, the focus for the PTF is on the core scenarios that relate to the kinds of users who will engage with TTS.
496 |User scenarios, like Personas, represent a composite of real-world experiences. In the case of the PTF, the scenarios were derived from interviews of people who were end-consumers of TTS, as well as submitted narratives and industry examples from practitioners. There are several formats of scenarios. Several are general goal or task-oriented scenarios. Others elaborate on richer context, for example, educational assessment.
497 |The following user scenarios are organized on the three perspectives of TTS use derived from analysis of the qualitative data collected from the discovery work:
498 |Need to add the other categories, or remove the list above and just rely on the ToC.
504 |As an AAC User I want my name to be pronounced correctly and I want to pronounce others names correctly using my AAC device.
511 |As an AAC user, I want to be able to input and store the correct pronunciation of others’ names, so I can address people respectfully and build meaningful relationships.
514 |For instance, when meeting someone named “Nguyễn,” the AAC user wants to ensure their device pronounces the name correctly, using IPA or SSML markup, to foster respectful communication and avoid embarrassment.
515 |As an AAC user, I want my name to be pronounced correctly by my device, so that I can confidently introduce myself in social, educational, and professional settings.
519 |For example, a user named “Siobhán” may find that default TTS engines mispronounce her name. She wants to input a phonetic or SSML-based pronunciation so that her name is spoken accurately every time.
520 |Ultimately, the quality and variation of TTS rendering by assistive technologies vary widely according to a user's context. The following user scenarios reinforce the necessity for accurate pronunciation from the perspective of those who consume digitally generated content.
527 |As a traveler who uses assistive technology (AT) with TTS to help navigate through websites, I need to hear arrival and destination codes pronounced accurately so I can select the desired travel itinerary. For example, a user with a visual impairment attempts to book a flight to Ottawa, Canada and so goes to a travel website. The user already knows the airport code and enters "YOW". The site produces the result in a drop-down list as "Ottawa, CA" but the AT does not pronounce the text accurately to help the user make the correct association between their data entry and the list item.
530 |As a test taker (tester) with a visual impairment who may use assistive technology to access the test content with speech software, screen reader or refreshable braille device, I want the content to be presented as intended, with accurate pronunciation and articulation, so that my assessment accurately reflects my knowledge of the content.
534 |As a student/learner with auditory and cognitive processing issues, it is difficult to distinguish sounds, inflections, and variations in pronunciation as rendered through synthetic voice, such as text-to-speech or screen reader technologies. Consistent and accurate pronunciation whether human-provided, external, or embedded is needed to support working executive processing, auditory processing and memory that facilitates comprehension in literacy and numeracy for learning and for assessments.
538 |As an English Learner (EL) or a visually impaired early learner using speech synthesis for reading comprehension that includes decoding words from letters as part of the learning construct (intent of measurement), pronunciation accuracy is vital to successful comprehension, as it allows the learner to distinguish sounds at the sentence, word, syllable, and phoneme level.
542 |The advent of graphical user interfaces (GUIs) for the management and editing of text content has given rise to content creators not requiring technical expertise beyond the ability to operate a text editing application such as Microsoft Word. The following scenario summarizes the general use, accompanied by a hypothetical application.
548 |In the educational assessment field, providing accurate and concise pronunciation for students with auditory accommodations, such as text-to-speech (TTS) or students with screen readers, is vital for ensuring content validity and alignment with the intended construct, which objectively measures a test takers knowledge and skills. For test administrators/educators, pronunciations must be consistent across instruction and assessment in order to avoid test bias or impact effects for students. Some additional requirements for the test administrators, include, but are not limited to, such scenarios:
556 |As a test administrator, I want to ensure that students with the read-aloud accommodation, who are using assistive technology or speech synthesis as an alternative to a human reader, have the same speech quality (e.g., intonation, expression, pronunciation, and pace, etc.) as a spoken language.
559 |This may be simlar to the other Test Administrator case below?
560 |As a math educator, I want to ensure that speech accuracy with mathematical expressions, including numbers, fractions, and operations have accurate pronunciation for those who rely on TTS. Some mathematical expressions require special pronunciations to ensure accurate interpretation while maintaining test validity and construct. Specific examples include:
564 |Mathematical formulas written in simple text with special formatting should convey the correct meaning of the expression to identify changes from normal text to super- or to sub-script text. For example, without the proper formatting, the equation:a3-b3=(a-b)(a2+ab+b2) may incorrectly render through some technologies and applications as a3-b3=(a-b)(a2+ab+b2).
Distinctions made in writing are often not made explicit in speech; For example, “fx” may be interpreted as fx, f(x), fx, F X, F X. The distinction depends on the context; requiring the author to provide consistent and accurate semantic markup.
571 |For math equations with Greek letters, it is important that the speech synthesizer be able to distinguish the phonetic differences between them, whether in the natural language or phonetic equivalents. For example, ε (epsilon) υ (upsilon) φ (phi) χ (chi) ξ(xi).
575 |As a test administrator/educator, pronunciations must be consistent across instruction and assessment, in order to avoid test bias and pronunciation effects on performance for students with disabilities (SWD) in comparison to students without disabilities (SWOD). Examples include:
580 |If a test question is measuring rhyming of words or sounds of words, the speech synthesis should not read aloud the words, but rather spell out the words in the answer options.
583 |If a test question is measuring spelling and the student needs to consider spelling correctness/incorrectness, the speech synthesis should not read aloud the misspelt words, especially for words, such as:
587 |The extension of content management in TTS is one as a means of encoding and preserving spoken text for academic analyses; irrespective of discipline, subject domain, or research methodology.
599 |A. As a linguist, I want to represent all the pronunciation variations of a given word in any language, for future analyses.
602 |As a speech language pathologist or speech therapists, I want TTS functionality to include components of speech and language that include dialectal and individual differences in pronunciation; identify differences in intonation, syntax, and semantics, and; allow for enhanced comprehension, language processing and support phonological awareness.
606 |Technical standards for software development assist organizations and individuals to provide accessible experiences for users with disabilities. The final user scenarios in this document are considered from the perspective of those who design and develop software.
613 |Probably shouldn't use "final" here, as we mey re-order.
614 |As a Product Owner for a web content management system (CMS), I want the next software product release to have the capability of pronouncing speech "just like Alexa can".
617 |As a client-side user interface developer, I need a way to render text content, so it is spoken accurately with assistive technologies.
621 |