Hacking

From FestivalTe

Jump to: navigation, search

Contents

Guide to Understanding festival-te

A description of the festival modules which were modified/written in Scheme for Telugu language are given here. Creating festival TTS for other languages, especially Indian languages is very similar and will mostly require language specific changes to be made to the code in these modules.

Modules

Text Normalization (telugu_scm/telugu_token.scm)

This module provides the rules to map tokens in an utterance to Telugu words. It defines the rules required to map numbers, dates, currencies, symbols, abbreviations etc. to words.

Phoneset Definition (telugu_scm/telugu_phones.scm)

This module defines the complete set of phones used in Telugu speech. It also includes feature definitions (ex. vowel/consonant, lip rounding) of these phones.

Any new voice built for Telugu language, should comply with this phoneset.

Lexical Analysis (telugu_scm/telugu_lex.scm)

Lexical analysis is used to arrive at the phones that make up the pronounciation of a particular word. Since Telugu is phonetic in nature, we do not require a dictionary for lexical analysis. Instead, this module defines letter-to-sound rules (lts) which are used to arrive at the speech phones based on the spelling of the word.

Prosody Modeling (telugu_scm/telugu_NSK_int.scm)

Prosodic phrasing in speech synthesis makes the whole speech more understandable. Phrasing is done based on punctuation. The module also defines accent and F0 prediction methods.


Telugu Diphone Voice

telugu_NSK_diphone voice is available for use with festival-te. It consists of a voice definition given in FESTIVAL_DIR/voices/telugu/festvox/telugu_NSK_diphone.scm and a diphone database located at FESTIVAL_DIR/voices/telugu/group/NSKlpc.group


Using Telugu Diphone Voice for Other Languages

The voice database NSKlpc.group is a generic voice, not specific to Telugu. The phoneset covers most Indian languages, so it can very well be used for other Indian languages. Please contact DONLab, Dept. of CS&E, IIT Madras for queries on using it for other Indian languages.

Personal tools