mirror of
https://github.com/rhasspy/espeak-ng.git
synced 2026-10-31 13:38:47 +00:00
jonsd
e69ec3a676
lang=en-r: change to phoneme [t#] (reduced "t") Added "consonants" attribute to voice files. Minor change to intonation types 2 and 3. lang=zh: adjust tone levels lang=fr: adjust stressAmp, adjust consonant strengths. Platform RISC_OS. Disable the "breath" attribute in voices because floating point is not allowed in callbacks. Platform RISC_OS. Implement ListVoices() git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@132 d46cf337-b52f-0410-862d-fd96e6ae7743
These are the spelling-to-phoneme pronunciation rules and exceptions lists for each language. You can make changes and then re-compile them to produce the "_dict" file in the espeak-data data directory (eg. "en_dict") which the speak program used. Use: speak --compile=<voicename> to compile the data for a particular language. It automatically replaces the corresponding _dict file in espeak-data, ready for use by the speak program. An "_extra" file, (eg. en_extra) can be used to in addition to the "_list" file to add exception words without disturbing the official "_list" file. It is treated as an extension of the "_list" file. Language Notes: Russian. An additional dictionary of Russian words and their stress positions is available at http://espeak.sourceforge.net/data/russian_data.zip. This allows eSpeak to put the stress on the correct syllable of Russian words.