Views
No views yet
sherpa-onnx/)| File | Role |
|---|---|
twi_ep045_steps4.onnx | acoustic model — default, 4 ODE steps, best quality |
twi_ep045_steps2.onnx | acoustic model — fast, 2 ODE steps, lower quality |
vocos-22khz-univ.onnx | Vocos universal vocoder |
tokens.txt | phoneme → id table |
espeak-ng-data/ | espeak-ng lfn-voice data (phonemization) |
sherpa-onnx)1pip install -U sherpa-onnx
2curl -L -o nano-twi.zip https://github.com/michsethowusu/nano-twi/releases/download/v1.0/nano-twi-sherpa-onnx.zip
3unzip nano-twi.zip
4curl -O https://raw.githubusercontent.com/michsethowusu/nano-twi/main/examples/python/synthesize.py
5python3 synthesize.py --model-dir ./nano-twi \
6 --text "Awurade ne me hwɛfoɔ, biribiara renhia me." --out twi.wav
7# fast model: add --acoustic twi_ep045_steps2.onnxhf download ghananlpcommunity/nano-twi --include "sherpa-onnx/*".)ɛ ɔ); phonemized to IPA via espeak-ng lfn.espeak-ng-data/ as-is (built with eSpeak NG 1.52.0) and sherpa-onnx ≥ 1.12;
swapping in different espeak data can change pronunciation.lfn (Lingua Franca Nova) voice (which maps Twi orthography to clean IPA).