Pocket TTS
Collection
Kyutai Pocket TTS models in ONNX for phoonnx. โข 6 items โข Updated
How to use Phoonnx/pocket-tts-german with Pocket-TTS:
from pocket_tts import TTSModel
import scipy.io.wavfile
tts_model = TTSModel.load_model("Phoonnx/pocket-tts-german")
voice_state = tts_model.get_state_for_audio_prompt(
"hf://kyutai/tts-voices/alba-mackenna/casual.wav"
)
audio = tts_model.generate_audio(voice_state, "Hello world, this is a test.")
# Audio is a 1D torch tensor containing PCM data.
scipy.io.wavfile.write("output.wav", tts_model.sample_rate, audio.numpy())A Pocket TTS model by Kyutai, from kyutai/pocket-tts/languages/german, packaged for phoonnx, our ONNX text-to-speech library.
| Architecture | Pocket TTS |
| Upstream | kyutai/pocket-tts/languages/german |
| ONNX conversion | KevinAHM/pocket-tts-onnx/onnx/german |
| Author | Kyutai |
| Languages | de |
| Text input | Unicode codepoints of the text |
pip install --pre phoonnx
import wave
from phoonnx.model_manager import TTSModelManager
manager = TTSModelManager()
manager.load()
manager.merge_default_voices()
voice = manager.voices["pockettts/de/alba"].load()
with wave.open("output.wav", "wb") as wav_file:
text = "..." # a sentence in de
voice.synthesize_wav(text, wav_file)
In OVOS, phoonnx installs the ovos-tts-plugin-phoonnx plugin. Select the voice in mycroft.conf:
{"tts": {"module": "ovos-tts-plugin-phoonnx", "ovos-tts-plugin-phoonnx": {"voice": "pockettts/de/alba"}}}
phoonnx hands the text to scriptconv, our text conversion library, which normalises it and turns it into the model's input:
unicode: UnicodeCodepointPhonemizer from scriptconv.phonemizers.base, in scriptconv itself.pockettts/de/albapockettts/de/annapockettts/de/azelmapockettts/de/bill_boerstpockettts/de/caro_davypockettts/de/charlespockettts/de/cosettepockettts/de/eponinepockettts/de/estellepockettts/de/evepockettts/de/fantinepockettts/de/georgepockettts/de/giovannipockettts/de/janepockettts/de/javertpockettts/de/jeanpockettts/de/juergenpockettts/de/lolapockettts/de/mariuspockettts/de/marypockettts/de/michaelpockettts/de/paulpockettts/de/peter_yearsleypockettts/de/rafaelpockettts/de/stuart_bellpockettts/de/veraThe upstream declares the licence cc-by-4.0 in the model card of kyutai/pocket-tts.