Parler-TTS: fully open-source high-quality TTS

parler-tts 's Collections

updated Aug 8

If you want to find out more about how these models were trained and even fine-tune them yourself, check-out the Parler-TTS repository on GitHub.

Running on Zero

731

🥖

Parler-TTS

High-fidelity Text-To-Speech
parler-tts/parler-tts-large-v1

Text-to-Speech • Updated Aug 8 • 17.2k • 201

Note Parler-TTS Large is a 2.2B-parameters Parler checkpoint, trained on 45K hours of audio data.
parler-tts/parler-tts-mini-v1

Text-to-Speech • Updated Aug 8 • 36.7k • 114

Note Parler-TTS Mini is a 880M parameters Parler checkpoint, trained on 45K hours of audio data.
Natural language guidance of high-fidelity text-to-speech with synthetic annotations

Paper • 2402.01912 • Published Feb 2 • 11
parler-tts/parler_tts_mini_v0.1

Text-to-Speech • Updated Apr 30 • 26.3k • 346

Note Parler-TTS v0.1 is a lightweight text-to-speech (TTS) model, trained on 10.5K hours of audio data, that can generate high-quality, natural sounding speech with features that can be controlled using a simple text prompt (e.g. gender, background noise, speaking rate, pitch and reverberation). It is the first release model from the Parler-TTS project, which aims to provide the community with TTS training resources and dataset pre-processing code. V1 coming soon!
parler-tts/dac_44khZ_8kbps

Updated Apr 10 • 1.46k • 15

Note Used to recover the audio waveform from the audio tokens predicted by the decoder. We use the DAC model from Descript, although other codec models, such as EnCodec, can also be used.
google/flan-t5-base

Text2Text Generation • Updated Jul 17, 2023 • 524k • • 792

Note Used to encode text descriptions.