Parakeet

Parakeet aims to provide a flexible, efficient and state-of-the-art text-to-speech toolkit for the open-source community. It is built on PaddlePaddle dynamic graph and includes many influential TTS models.

News

• Aug-31-2021, Chinese Text Frontend. Check examples/text_frontend.
• Aug-23-2021, FastSpeech2 with AISHELL-3. Check fastspeech2/aishell3.
• Aug-3-2021, FastSpeech2 with CSMSC. Check fastspeech2/baker.
• Jul-19-2021, SpeedySpeech with CSMSC. Check speedyspeech/baker.
• Jul-01-2021, Parallel WaveGAN with CSMSC. Check parallelwave_gan/baker.
• Jul-01-2021, Montreal-Forced-Aligner. Check examples/use_mfa.
• May-07-2021, voice cloning in Chinese. Check examples/tacotron2_aishell3.

Overview

In order to facilitate exploiting the existing TTS models directly and developing the new ones, Parakeet selects typical models and provides their reference implementations in PaddlePaddle. Further more, Parakeet abstracts the TTS pipeline and standardizes the procedure of data preprocessing, common modules sharing, model configuration, and the process of training and synthesis. The models supported here include Text FrontEnd, end-to-end Acoustic models and Vocoders:

• Text FrontEnd
  • Rule based frontend.

• Acoustic Models
  • 【FastSpeech2】FastSpeech 2: Fast and High-Quality End-to-End Text to Speech
  • 【SpeedySpeech】SpeedySpeech: Efficient Neural Speech Synthesis
  • 【Transformer TTS】Neural Speech Synthesis with Transformer Network
  • 【Tacotron2】Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
• Vocoders
  • 【Parallel WaveGAN】Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram
  • 【WaveFlow】WaveFlow: A Compact Flow-based Model for Raw Audio
• Voice Cloning
  • Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesis
  • 【GE2E】Generalized End-to-End Loss for Speaker Verification

Setup

It's difficult to install some dependent libraries for this repo in Windows system, we recommend that you DO NOT use Windows system, please use Linux.

Make sure the library libsndfile1 is installed, e.g., on Ubuntu.

sudo apt-get install libsndfile1

Install PaddlePaddle

See install for more details. This repo requires PaddlePaddle 2.1.2 or above.

Install Parakeet

pip install -U paddle-parakeet

or

git clone https://github.com/PaddlePaddle/Parakeet
cd Parakeet
pip install -e .

If some python dependent packages cannot be installed successfully, you can run the following script first. (replace python3.6 with your own python version)

sudo apt install -y python3.6-dev

See install for more details.

Examples

Entries to the introduction, and the launch of training and synthsis for different example models:

• >>> Chinese Text Frontend
• >>> FastSpeech2
• >>> Montreal-Forced-Aligner
• >>> Parallel WaveGAN
• >>> SpeedySpeech
• >>> Tacotron2_AISHELL3
• >>> GE2E
• >>> WaveFlow
• >>> TransformerTTS
• >>> Tacotron2

Audio samples

TTS models (Acoustic Model + Neural Vocoder)

Check our website for audio sampels.

Released Model

AM

FastSpeech2

1. fastspeech2_nosil_baker_ckpt_0.4.zip
2. fastspeech2_nosil_aishell3_ckpt_0.4.zip

SpeedySpeech

1. speedyspeech_baker_ckpt_0.4.zip

TransformerTTS

1. transformer_tts_ljspeech_ckpt_0.3.zip

Tacotron2

1. tacotron2_ljspeech_ckpt_0.3.zip
2. tacotron2_ljspeech_ckpt_0.3_alternative.zip

Vocoder

WaveFlow

1. waveflow_ljspeech_ckpt_0.3.zip

Parallel WaveGAN

1. pwg_baker_ckpt_0.4.zip

Voice Cloning

Tacotron2_AISHELL3

1. tacotron2_aishell3_ckpt_0.3.zip

GE2E

1. ge2e_ckpt_0.3.zip

License

Parakeet is provided under the Apache-2.0 license.