Requirements in PLS, SSML and SRGS Standard—Hindi as a Case Study
| Autor: | Prashant Verma, Bharat Gupta, Vijay Kumar |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Hindi
computer.internet_protocol Computer science business.industry Speech synthesis Romance languages Pronunciation computer.software_genre language.human_language language Speech Synthesis Markup Language Artificial intelligence Speech Recognition Grammar Specification Pronunciation Lexicon Specification business computer XML Natural language processing |
| Zdroj: | Smart Systems: Innovations in Computing ISBN: 9789811628764 |
| DOI: | 10.1007/978-981-16-2877-1_12 |
| Popis: | The Speech Synthesis Markup Language (SSML) specification is one of the speech standard of W3C (World-Wide Web Consortium) and is aimed to provide a rich, XML (Extensible Markup Language) for assisting the generation of synthetic speech on Web and other applications. It enables authors a standard way to control various speech parameters in order to improve the quality and give natural voice to the speech synthesis applications such as TTS. Speech Recognition Grammar Specification (SRGS) is a W3C standard that enables developers to define the words and patterns of words that recognized by speech recognizer. W3C PLS (Pronunciation Lexicon Specification) represents the pronunciation information of the words for use in speech recognition and synthesis systems. Currently, these standards enable the output in a large numbers of Latin languages only. Therefore, it is essential to study the existing standards for Indian languages as Indian languages are morphologically rich languages. This paper presents the various aspects of these technologies that need to be identified for Indian languages. The papers proposed the some of the identified requirements/gaps in these speech standards. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|