1. PLS for SSML Paolo Baggia Loquendo Workshop II on Internationalizing SSML

2. 2 Outline Brief Introduction of Pronunciation Lexicon Specification Examples of common use cases How to deal of Homographs in PLS Other issues for the workshop

3. 3 Why Pronunciation Lexicon Specification? Allow to customize pronunciations (for proper names, locations, addresses, acronyms, etc.) Enrich TTS and ASR with customized pronunciations Complete the “Speech Interaction Framework”  Read the specification at: http://www.w3.org/TR/pronunciation-lexicon/

4. 4 What PLS 1.0 is Not! Multilingual pronunciation lexicon  the current specification is mono-lingual! Extended purpose lexicon  no syntax, no semantics, no morphology TTS-internal lexicon  too complex and rich of custom knowledge PLS 1.0 is restricted to the most important and tractable issues.

5. 5 The PLS 1.0 Language PLS is an XML language The root element is, with attributes version, xmlns, alphabet, xml:lang It contains a collection s, which is composed of: – s for orthographies/spellings – s for pronunciations – s for textual substitutions – s for examples  The order is relevant to determine the preferred pronunciation for TTS

6. 6 PLS Common Usages Multiple pronunciations –For ASR: to accommodate speaker/regional variability, not native speakers –For TTS: a single preferred pronunciation will be selected Multiple orthographies (with same pronunciations) –For ASR & TTS: Homophones (same pronunciations, different meanings) –Different s Homographs (same spellings, different pronunciations) –This creates troubles  we propose a solution

7. 7 How to Deal with Homographs New attribute “ role ” on elements Values are qnames (qualified names, with a namespace) e.g. “myvocabulary:verb”, “wordnet:verb” Open to future standardization  allows both proprietary values and standard ones More than one qname for a single, e.g. role=“w:verb w:past-tense”

8. 8 Example of Homographs in PLS <lexicon version="1.0" xmlns="http://www.w3.org/2005/pronunciation-lexicon"http://www.w3.org/2005/pronunciation-lexicon alphabet="ipa" xml:lang="en-GB"> refuse r ɪ 'fju:z refuse 'refju:s Example sentence: “I refuse to take the fridge as a refuse.”

9. 9 SSML should to be extended! I refuse to take the fridge as a refuse. … and SRGS too!

10. 10 Other Issues for the Workshop Lexicon selection criteria in SSML Which phonetic alphabets?  Current PLS 1.0 mandates the usage of IPA (International Pronunciation Alphabet)  Are there other options? We need standard alphabets! Other issues?