What changed
1 fix1 addition4 changes1 removal
changedPocketTTS CPU voice cloning now supports French, German, Spanish, and Italian alongside English. The required multilingual models are included, and PocketTTS automatically uses the appropriate existing voice reference for the selected language when a personal reference has not been provided.
changedPocketTTS settings and language information have been updated across all supported app languages.
changedTTS Audio Export
addedA new TTS Audio Export tool is available at the bottom of Advanced Settings.
changedText can be entered directly or imported from a text file, then converted into WAV files using the currently active TTS engine, language, voice, and settings. Content can be divided by paragraphs, lines, or sentences, with multiple sections optionally combined into each audio file. Longer documents are processed sequentially using the same established reading and speech flow used elsewhere in the app.
removedCompleted audio is saved inside a timestamped folder under StreamingAssets/GeneratedTTS. The output folder opens automatically when generation finishes. Generation can be cancelled without removing files that have already been completed.
CielChan: Anime Desktop AI Companion changes
changedPocketTTS CPU voice cloning now supports French, German, Spanish, and Italian alongside English. The required multilingual models are included, and PocketTTS automatically uses the appropriate existing voice reference for the selected language when a personal reference has not been provided.
changedPocketTTS settings and language information have been updated across all supported app languages.
changedTTS Audio Export
addedA new TTS Audio Export tool is available at the bottom of Advanced Settings.
changedText can be entered directly or imported from a text file, then converted into WAV files using the currently active TTS engine, language, voice, and settings. Content can be divided by paragraphs, lines, or sentences, with multiple sections optionally combined into each audio file. Longer documents are processed sequentially using the same established reading and speech flow used elsewhere in the app.
Hi everyone,
v1.7.9 has been released:
Multilingual PocketTTS
PocketTTS CPU voice cloning now supports French, German, Spanish, and Italian alongside English. The required multilingual models are included, and PocketTTS automatically uses the appropriate existing voice reference for the selected language when a personal reference has not been provided.
PocketTTS settings and language information have been updated across all supported app languages.
TTS Audio Export
A new TTS Audio Export tool is available at the bottom of Advanced Settings.
Text can be entered directly or imported from a text file, then converted into WAV files using the currently active TTS engine, language, voice, and settings. Content can be divided by paragraphs, lines, or sentences, with multiple sections optionally combined into each audio file. Longer documents are processed sequentially using the same established reading and speech flow used elsewhere in the app.
Completed audio is saved inside a timestamped folder under StreamingAssets/GeneratedTTS. The output folder opens automatically when generation finishes. Generation can be cancelled without removing files that have already been completed.
Small Bug Fixes
Various additional PocketTTS startup, multilingual generation, full-response playback, file-import, localization, and interface reliability improvements. Fixed an issue affecting certain languages when using Kokoro in v1.7.8.
More to come soon.
