mixertts-ljspeech-80

A MixerTTS model by nipponjo, from github.com/nipponjo/mixer-tts-pytorch, packaged for phoonnx, our ONNX text-to-speech library.

Model details

Architecture	MixerTTS
Upstream	github.com/nipponjo/mixer-tts-pytorch
Author	nipponjo
Languages	en-US
Text input	espeak phonemes, IPA alphabet
Vocoder	vocoders-vocos-mel-22khz-univ

How to use

pip install --pre phoonnx

The phonemes come from the espeak-ng program, so install it too (for example apt install espeak-ng). Without it, pip install --pre "phoonnx[espeak]" gives phoonnx the espyak port of espeak-ng. Some espyak phonemes differ from those of espeak-ng, so the voice can say some words wrong with espyak.

import wave
from phoonnx.model_manager import TTSModelManager

manager = TTSModelManager()
manager.load()
manager.merge_default_voices()

voice = manager.voices["nipponjo/mixer-tts-ljspeech-80"].load()

with wave.open("output.wav", "wb") as wav_file:
    text = "..."  # a sentence in en-US
    voice.synthesize_wav(text, wav_file)

The model makes a spectrogram, and the vocoder in Model details turns it into audio. phoonnx downloads the vocoder with the voice.

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": "nipponjo/mixer-tts-ljspeech-80"}}}

Text front end

phoonnx hands the text to scriptconv, our text conversion library, which normalises it and turns it into the model's input:

• espeak: EspeakPhonemizer from scriptconv.phonemizers.mul, installed by the scriptconv[espeak] extra.

phoonnx voices

• nipponjo/mixer-tts-ljspeech-80

Licence

No licence is recorded for this model. Check the upstream release before you use it.

Related projects

• phoonnx: runs this model, in Python and in OVOS.
• scriptconv: text normalisation, script conversion and phonemizers for phoonnx.
• The MixerTTS collection: every MixerTTS model packaged for phoonnx.