I just found this. Main page [https://k2-fsa.github.io/sherpa/onnx/index.html]
This is huge! As a german, I use thorsten medium
[https://huggingface.co/csukuangfj/sherpa-onnx-apk/resolve/main/tts-engine-new/1.10.26/sherpa-onnx-1.10.26-arm64-v8a-de-tts-engine-vits-piper-de_DE-thorsten-medium.apk]
as he simply made the best dataset. Mixing english with german, speaking
numbers, single letters, pausing without a “.” but just a linebreak, all those
can be essential. And… it is nearly perfect! And all local! This is crazy!
eSpeak can finally go to rest!
I’ve been using this for a few months now with Organic Maps and it’s incredibly good. A few versions ago I noticed some issues with certain letter sounds but with frequent updates it now sounds so natural.
I found “vits-piper-en_GB-southern_english_female-medium” to be the most natural sounding voice for English.
I’ve been using this for a few months now with Organic Maps and it’s incredibly good. A few versions ago I noticed some issues with certain letter sounds but with frequent updates it now sounds so natural.
I found “vits-piper-en_GB-southern_english_female-medium” to be the most natural sounding voice for English.
Nice! It all depends on the model.
Thorsten has spoken 23+h of audios into his main dataset, crazy