There is a newer version of the record available.

Published April 15, 2021 | Version v0.0.12
Software Open

coqui-ai/TTS: v0.0.12

Creators

  • 1. Coqui.ai
  • 2. University of São Paulo (USP)
  • 3. Faculty of Electrical Engineering and Information Technologies
  • 4. Airbus Defence & Space
  • 5. no
  • 6. Inveniem
  • 7. Aalto University
  • 8. @coqui-ai

Description

:frog: v0.0.12 🐞Bug Fixes
  • [x] fix #419 (This is a crucial bug fix).
  • [x] fix #408
💾 Code updates
  • [x] Enable logging model config.json on Tensorboard. #418
  • [x] Update code style standards and use a Makefile to ease regular tasks. #423
  • [x] Enable using Tacotron.prenet.dropout at inference time. This leads to a better quality with some models.
  • [x] Update default tts model to LJspeech TacotronDDC.
  • [x] Show the real waveform on Tensorboard in GAN vocoder training.
:walking_woman: Operational Updates 🏅 Model implementations
  • [x] initial HiFiGAN implementation (:crown: @rishikksh20 @erogol) #422
🚀 New Pre-Trained Model Releases
  • [ ] Universal HifiGAN model(postponed to the next version for :crown: @Edresson's updated model.)
  • [x] LJSpeech, Tacotron2 Double Decoder Consistency v2 model.
     Check our [blog post](https://coqui.ai/blog/tts/solving-attention-problems-of-tts-models-with-double-decoder-consistency) to learn more about Double Decoder Consistency.
  • [x] LJSpeech HifiGAN model.
Released Models

:bulb: All the models below are available by tts end point as explained here.

Language Dataset Model Name Model Type TTS version Download

:sparkles: English | LJSpeech | Tacotron-DDC | tts |:smiley: v0.0.12| :floppy_disk: German | Thorsten-DE | Tacotron-DCA | tts | v0.0.11 |:floppy_disk: German | Thorsten-DE | Wavegrad | vocoder |v0.0.11 |:floppy_disk: English | LJSpeech | SpeedySpeech | tts | v0.0.10 |:floppy_disk: English | EK1 | Tacotron2 | tts |v0.0.10 |:floppy_disk: Russian | RUSLAN | TacotronDDC | tts | v0.0.10 |:floppy_disk: Dutch | MAI | TacotronDDC | tts | v0.0.10 |:floppy_disk: Chinese | Baker | TacotronDDC-GST | tts | v0.0.10 |:floppy_disk: English | LJSpeech | TacotronDCA | tts |v0.0.9 |:floppy_disk: English | LJSpeech | Glow-TTS | tts |v0.0.9 |:floppy_disk: Spanish | M-AILabs | TacotronDDC | tts |v0.0.9 |:floppy_disk: French | M_AILabs | TacotronDDC | tts |v0.0.9 |:floppy_disk: Dutch | MAI | TacotronDDC | tts | v0.0.10 |:floppy_disk: :sparkles: English | LJSpeech | HiFiGAN | vocoder | :smiley: v0.0.12| :floppy_disk: English | EK1 | WaveGrad | vocoder | v0.0.10 |:floppy_disk: Dutch | MAI | ParallelWaveGAN | vocoder | v0.0.10 |:floppy_disk: English | LJSpeech | MB-MelGAN | vocoder |v0.0.9 |:floppy_disk: :earth_africa: Multi-Lang | LibriTTS | FullBand-MelGAN | vocoder |v0.0.9 |:floppy_disk: :earth_africa: Multi-Lang | LibriTTS | WaveGrad | vocoder |v0.0.9 |:floppy_disk:

Files

coqui-ai/TTS-v0.0.12.zip

Files (12.1 MB)

Name Size Download all
md5:8e4cf2874a0b4dfa99e839a9e211970a
12.1 MB Preview Download

Additional details

Related works

Is supplement to
https://github.com/coqui-ai/TTS/tree/v0.0.12 (URL)