PyTorch implementation of Microsoft's text-to-speech system FastSpeech 2: Fast and High-Quality End-to-End Text to Speech.

Overview

FastSpeech 2 - PyTorch Implementation

This is a PyTorch implementation of Microsoft's text-to-speech system FastSpeech 2: Fast and High-Quality End-to-End Text to Speech. This project is based on xcmyz's implementation of FastSpeech. Feel free to use/modify the code.

There are several versions of FastSpeech 2. This implementation is more similar to version 1, which uses F0 values as the pitch features. On the other hand, pitch spectrograms extracted by continuous wavelet transform are used as the pitch features in the later versions.

Updates

  • 2021/7/8: Release the checkpoint and audio samples of a multi-speaker English TTS model trained on LibriTTS
  • 2021/2/26: Support English and Mandarin TTS
  • 2021/2/26: Support multi-speaker TTS (AISHELL-3 and LibriTTS)
  • 2021/2/26: Support MelGAN and HiFi-GAN vocoder

Audio Samples

Audio samples generated by this implementation can be found here.

Quickstart

Dependencies

You can install the Python dependencies with

pip3 install -r requirements.txt

Inference

You have to download the pretrained models and put them in output/ckpt/LJSpeech/, output/ckpt/AISHELL3, or output/ckpt/LibriTTS/.

For English single-speaker TTS, run

python3 synthesize.py --text "YOUR_DESIRED_TEXT" --restore_step 900000 --mode single -p config/LJSpeech/preprocess.yaml -m config/LJSpeech/model.yaml -t config/LJSpeech/train.yaml

For Mandarin multi-speaker TTS, try

python3 synthesize.py --text "大家好" --speaker_id SPEAKER_ID --restore_step 600000 --mode single -p config/AISHELL3/preprocess.yaml -m config/AISHELL3/model.yaml -t config/AISHELL3/train.yaml

For English multi-speaker TTS, run

python3 synthesize.py --text "YOUR_DESIRED_TEXT"  --speaker_id SPEAKER_ID --restore_step 800000 --mode single -p config/LibriTTS/preprocess.yaml -m config/LibriTTS/model.yaml -t config/LibriTTS/train.yaml

The generated utterances will be put in output/result/.

Here is an example of synthesized mel-spectrogram of the sentence "Printing, in the only sense with which we are at present concerned, differs from most if not from all the arts and crafts represented in the Exhibition", with the English single-speaker TTS model.

Batch Inference

Batch inference is also supported, try

python3 synthesize.py --source preprocessed_data/LJSpeech/val.txt --restore_step 900000 --mode batch -p config/LJSpeech/preprocess.yaml -m config/LJSpeech/model.yaml -t config/LJSpeech/train.yaml

to synthesize all utterances in preprocessed_data/LJSpeech/val.txt

Controllability

The pitch/volume/speaking rate of the synthesized utterances can be controlled by specifying the desired pitch/energy/duration ratios. For example, one can increase the speaking rate by 20 % and decrease the volume by 20 % by

python3 synthesize.py --text "YOUR_DESIRED_TEXT" --restore_step 900000 --mode single -p config/LJSpeech/preprocess.yaml -m config/LJSpeech/model.yaml -t config/LJSpeech/train.yaml --duration_control 0.8 --energy_control 0.8

Training

Datasets

The supported datasets are

  • LJSpeech: a single-speaker English dataset consists of 13100 short audio clips of a female speaker reading passages from 7 non-fiction books, approximately 24 hours in total.
  • AISHELL-3: a Mandarin TTS dataset with 218 male and female speakers, roughly 85 hours in total.
  • LibriTTS: a multi-speaker English dataset containing 585 hours of speech by 2456 speakers.

We take LJSpeech as an example hereafter.

Preprocessing

First, run

python3 prepare_align.py config/LJSpeech/preprocess.yaml

for some preparations.

As described in the paper, Montreal Forced Aligner (MFA) is used to obtain the alignments between the utterances and the phoneme sequences. Alignments of the supported datasets are provided here. You have to unzip the files in preprocessed_data/LJSpeech/TextGrid/.

After that, run the preprocessing script by

python3 preprocess.py config/LJSpeech/preprocess.yaml

Alternately, you can align the corpus by yourself. Download the official MFA package and run

./montreal-forced-aligner/bin/mfa_align raw_data/LJSpeech/ lexicon/librispeech-lexicon.txt english preprocessed_data/LJSpeech

or

./montreal-forced-aligner/bin/mfa_train_and_align raw_data/LJSpeech/ lexicon/librispeech-lexicon.txt preprocessed_data/LJSpeech

to align the corpus and then run the preprocessing script.

python3 preprocess.py config/LJSpeech/preprocess.yaml

Training

Train your model with

python3 train.py -p config/LJSpeech/preprocess.yaml -m config/LJSpeech/model.yaml -t config/LJSpeech/train.yaml

The model takes less than 10k steps (less than 1 hour on my GTX1080Ti GPU) of training to generate audio samples with acceptable quality, which is much more efficient than the autoregressive models such as Tacotron2.

TensorBoard

Use

tensorboard --logdir output/log/LJSpeech

to serve TensorBoard on your localhost. The loss curves, synthesized mel-spectrograms, and audios are shown.

Implementation Issues

  • Following xcmyz's implementation, I use an additional Tacotron-2-styled Post-Net after the decoder, which is not used in the original FastSpeech 2.
  • Gradient clipping is used in the training.
  • In my experience, using phoneme-level pitch and energy prediction instead of frame-level prediction results in much better prosody, and normalizing the pitch and energy features also helps. Please refer to config/README.md for more details.

Please inform me if you find any mistakes in this repo, or any useful tips to train the FastSpeech 2 model.

References

Citation

@INPROCEEDINGS{chien2021investigating,
  author={Chien, Chung-Ming and Lin, Jheng-Hao and Huang, Chien-yu and Hsu, Po-chun and Lee, Hung-yi},
  booktitle={ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, 
  title={Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech}, 
  year={2021},
  volume={},
  number={},
  pages={8588-8592},
  doi={10.1109/ICASSP39728.2021.9413880}}
Owner
Chung-Ming Chien
Graduate Student, NTU CSIE | Speech Processing Lab. | Speech synthesis & Natural language processing
Chung-Ming Chien
Rhasspy 673 Dec 28, 2022
DziriBERT: a Pre-trained Language Model for the Algerian Dialect

DziriBERT is the first Transformer-based Language Model that has been pre-trained specifically for the Algerian Dialect.

117 Jan 07, 2023
LightSeq: A High-Performance Inference Library for Sequence Processing and Generation

LightSeq is a high performance inference library for sequence processing and generation implemented in CUDA. It enables highly efficient computation of modern NLP models such as BERT, GPT2, Transform

Bytedance Inc. 2.5k Jan 03, 2023
NVDA, the free and open source Screen Reader for Microsoft Windows

NVDA NVDA (NonVisual Desktop Access) is a free, open source screen reader for Microsoft Windows. It is developed by NV Access in collaboration with a

NV Access 1.6k Jan 07, 2023
Synthetic data for the people.

zpy: Synthetic data in Blender. Website • Install • Docs • Examples • CLI • Contribute • Licence Abstract Collecting, labeling, and cleaning data for

Zumo Labs 253 Dec 21, 2022
A python wrapper around the ZPar parser for English.

NOTE This project is no longer under active development since there are now really nice pure Python parsers such as Stanza and Spacy. The repository w

ETS 49 Sep 12, 2022
An open source library for deep learning end-to-end dialog systems and chatbots.

DeepPavlov is an open-source conversational AI library built on TensorFlow, Keras and PyTorch. DeepPavlov is designed for development of production re

Neural Networks and Deep Learning lab, MIPT 6k Dec 30, 2022
Disfl-QA: A Benchmark Dataset for Understanding Disfluencies in Question Answering

Disfl-QA is a targeted dataset for contextual disfluencies in an information seeking setting, namely question answering over Wikipedia passages. Disfl-QA builds upon the SQuAD-v2 (Rajpurkar et al., 2

Google Research Datasets 52 Jun 21, 2022
Large-scale Knowledge Graph Construction with Prompting

Large-scale Knowledge Graph Construction with Prompting across tasks (predictive and generative), and modalities (language, image, vision + language, etc.)

ZJUNLP 161 Dec 28, 2022
An official implementation for "CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval"

The implementation of paper CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval. CLIP4Clip is a video-text retrieval model based

ArrowLuo 456 Jan 06, 2023
The Internet Archive Research Assistant - Daily search Internet Archive for new items matching your keywords

The Internet Archive Research Assistant - Daily search Internet Archive for new items matching your keywords

Kay Savetz 60 Dec 25, 2022
Pytorch implementation of Tacotron

Tacotron-pytorch A pytorch implementation of Tacotron: A Fully End-to-End Text-To-Speech Synthesis Model. Requirements Install python 3 Install pytorc

soobin seo 203 Dec 02, 2022
Chinese Named Entity Recognization (BiLSTM with PyTorch)

BiLSTM-CRF for Name Entity Recognition PyTorch version A PyTorch implemention of Bi-LSTM-CRF model for Chinese Named Entity Recognition. 使用 PyTorch 实现

5 Jun 01, 2022
Model for recasing and repunctuating ASR transcripts

Recasing and punctuation model based on Bert Benoit Favre 2021 This system converts a sequence of lowercase tokens without punctuation to a sequence o

Benoit Favre 88 Dec 29, 2022
The first online catalogue for Arabic NLP datasets.

Masader The first online catalogue for Arabic NLP datasets. This catalogue contains 200 datasets with more than 25 metadata annotations for each datas

ARBML 94 Dec 26, 2022
This repository contains Python scripts for extracting linguistic features from Filipino texts.

Filipino Text Linguistic Feature Extractors This repository contains scripts for extracting linguistic features from Filipino texts. The scripts were

Joseph Imperial 1 Oct 05, 2021
Yet another Python binding for fastText

pyfasttext Warning! pyfasttext is no longer maintained: use the official Python binding from the fastText repository: https://github.com/facebookresea

Vincent Rasneur 230 Nov 16, 2022
Dense Passage Retriever - is a set of tools and models for open domain Q&A task.

Dense Passage Retrieval Dense Passage Retrieval (DPR) - is a set of tools and models for state-of-the-art open-domain Q&A research. It is based on the

Meta Research 1.1k Jan 07, 2023
多语言降噪预训练模型MBart的中文生成任务

mbart-chinese 基于mbart-large-cc25 的中文生成任务 Input source input: text + /s + lang_code target input: lang_code + text + /s Usage token_ids_mapping.jso

11 Sep 19, 2022