alphacep / vosk-tts

Text To Speech Synthesis with Vosk
Apache License 2.0
129 stars 18 forks source link

Turkish Language Support #8

Open trappedinspacetime opened 10 months ago

trappedinspacetime commented 10 months ago

Thank you for creating VOSK TTS models. Vosk has multilanguage support for STT inference. I tried Turkish STT model. It is very successful. I wish you add a Turkish TTS model.

nshmyrev commented 10 months ago

Eventually yes, we will have a Turkish model too. If you need a voice right now please let us know. Meanwhile you can try Piper voice probably, it supports Turkish https://rhasspy.github.io/piper-samples/

csukuangfj commented 10 months ago

Piper models are great.

By the way, we also provide C++ runtime support for all models from Piper, including the Turkish model. You can run it on macOS/Linux/Windows/Android/iOS/Raspberry Pi, etc. There are also a variety of programming language APIs that you can use, for instance, C++/C/Python/Kotlin/C#/Go/Java/Swift/Javascript, etc.

You can try the Turkish TTS model from within your browser without installing anything by visiting https://huggingface.co/spaces/k2-fsa/text-to-speech

Screenshot 2024-01-25 at 10 19 15


You can also try it on your phones. We have pre-built Android APKs available at https://k2-fsa.github.io/sherpa/onnx/tts/apk.html Screenshot 2024-01-25 at 10 21 02

If you want to replace your Android system TTS engine, then you can download APKs from https://k2-fsa.github.io/sherpa/onnx/tts/apk-engine.html


Everything is open-sourced. Please see https://github.com/k2-fsa/sherpa-onnx