SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based...
Transcript of SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based...
![Page 1: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/1.jpg)
SSML for Urdu Speech Synthesis
Sarmad Hussain
Professor and HeadCenter for Research in Urdu Language Processing
National University of Computer and Emerging Sciences, Lahore
¢
www.nu.edu.pk www.crulp.org
![Page 2: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/2.jpg)
SSML for Urdu TTS 2
Background
• Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode or UZT text and outputs Urdu
speech– performs textual analysis, normalization and
phonological processing to convert the text to speech– uses 5000+ diphone database– models duration and intonation based on learning
through tagged corpus– complete framework developed by CRULP through
funding by Ministry of IT, Govt. of Pakistan
![Page 3: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/3.jpg)
SSML for Urdu TTS 3
Using SSML for Urdu Speech Synthesis
• SSM L standards works well for Urdu• SSML needs to be enhanced to handle some
Urdu specific problems– Multilingual text– Digits– Date formats– Diacritics– Word Segmentation
![Page 4: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/4.jpg)
SSML for Urdu TTS 4
Multilingual Text
• Need to identify one of the three strategies (for different contexts)– Switch to English TTS– Transliterate and Read– Spell-out
![Page 5: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/5.jpg)
SSML for Urdu TTS 5
![Page 6: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/6.jpg)
SSML for Urdu TTS 6
![Page 7: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/7.jpg)
SSML for Urdu TTS 7
![Page 8: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/8.jpg)
SSML for Urdu TTS 8
![Page 9: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/9.jpg)
SSML for Urdu TTS 9
Digits
• Should be able to read out Urdu digits– Issues related to normalization?
• Two strategies for English digits (to be defined by the website)– Read in English– Read in Urdu
![Page 10: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/10.jpg)
SSML for Urdu TTS 10
Date Formats
• Cover other date formats in Urdu– A complete list can be provided
• Cover two calendars– Lunar based islamic calendar– Regular calendar
![Page 11: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/11.jpg)
SSML for Urdu TTS 11
![Page 12: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/12.jpg)
SSML for Urdu TTS 12
Diacritics
• Optionally, incompletely and (sometimes) incorrectly given
•ً
(almost), ٰا (high (in quality or rank))
• َِ ُ (ambiguous, knit (v), son of (particle), make (v))
![Page 13: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/13.jpg)
SSML for Urdu TTS 13
Diacritics• Lexical look-up
– Should be ignored, and look up base form from the lexicon– MUST match given diacritics (even if incomplete)
• Normally given where diacritic can cause ambiguity
• If lexical look-up fails, use a pronunciation guessing engine?
• Use default mechanism/ No engine• Use an engine, at the given path: need to decide on
– diacritics or pronunciation?» IPA, Custom Scheme, others?
– Accuracy vs. performance?» POS tagger» WSD
![Page 14: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/14.jpg)
SSML for Urdu TTS 14
Word Segmentation• Urdu does not have the concept of SPACE• SPACE used by typists to get the visual output
– Spaces within words• ور
– No spaces between words, esp. after non-joiners, as words visually OK
• ورڑزژ آدڈذا
• Should it point to word segmentation engine to enable correct output?– Use default mechanism/ No engine– Use an engine, at the given path: need to decide on
• Accuracy vs. performance?
![Page 15: SSML for Urdu TTS - World Wide Web Consortium · SSML for Urdu TTS 2 Background • Comments based on experience in developing a concatenative Urdu TTS system – inputs Urdu Unicode](https://reader030.fdocuments.us/reader030/viewer/2022020204/5add23d27f8b9a1a088cf549/html5/thumbnails/15.jpg)
SSML for Urdu Speech Synthesis
Sarmad Hussain
Professor and HeadCenter for Research in Urdu Language Processing
National University of Computer and Emerging Sciences, Lahore
¢
www.nu.edu.pk www.crulp.org