• Stars
    star
    6,749
  • Rank 5,784 (Top 0.2 %)
  • Language
    Jupyter Notebook
  • License
    Mozilla Public Li...
  • Created over 6 years ago
  • Updated over 1 year ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

๐Ÿค– ๐Ÿ’ฌ Deep learning for Text to Speech (Discussion forum: https://discourse.mozilla.org/c/tts)

TTS: Text-to-Speech for all.

TTS is a library for advanced Text-to-Speech generation. It's built on the latest research, was designed to achieve the best trade-off among ease-of-training, speed and quality. TTS comes with pretrained models, tools for measuring dataset quality and already used in 20+ languages for products and research projects.

CircleCI License PyPI version

๐Ÿ“ข English Voice Samples and SoundCloud playlist

๐Ÿ‘จโ€๐Ÿณ TTS training recipes

๐Ÿ“„ Text-to-Speech paper collection

๐Ÿ’ฌ Where to ask questions

Please use our dedicated channels for questions and discussion. Help is much more valuable if it's shared publicly, so that more people can benefit from it.

Type Platforms
๐Ÿšจ Bug Reports GitHub Issue Tracker
โ” FAQ TTS/Wiki
๐ŸŽ Feature Requests & Ideas GitHub Issue Tracker
๐Ÿ‘ฉโ€๐Ÿ’ป Usage Questions Discourse Forum
๐Ÿ—ฏ General Discussion Discourse Forum and Matrix Channel

๐Ÿ”— Links and Resources

Type Links
๐Ÿ’พ Installation TTS/README.md
๐Ÿ‘ฉ๐Ÿพโ€๐Ÿซ Tutorials and Examples TTS/Wiki
๐Ÿš€ Released Models TTS/Wiki
๐Ÿ’ป Docker Image Repository by @synesthesiam
๐Ÿ–ฅ๏ธ Demo Server TTS/server
๐Ÿค– Running TTS on Terminal TTS/README.md
โœจ How to contribute TTS/README.md

๐Ÿฅ‡ TTS Performance

"Mozilla*" and "Judy*" are our models. Details...

Features

  • High performance Deep Learning models for Text2Speech tasks.
    • Text2Spec models (Tacotron, Tacotron2, Glow-TTS, SpeedySpeech).
    • Speaker Encoder to compute speaker embeddings efficiently.
    • Vocoder models (MelGAN, Multiband-MelGAN, GAN-TTS, ParallelWaveGAN, WaveGrad, WaveRNN)
  • Fast and efficient model training.
  • Detailed training logs on console and Tensorboard.
  • Support for multi-speaker TTS.
  • Efficient Multi-GPUs training.
  • Ability to convert PyTorch models to Tensorflow 2.0 and TFLite for inference.
  • Released models in PyTorch, Tensorflow and TFLite.
  • Tools to curate Text2Speech datasets underdataset_analysis.
  • Demo server for model testing.
  • Notebooks for extensive model benchmarking.
  • Modular (but not too much) code base enabling easy testing for new ideas.

Implemented Models

Text-to-Spectrogram

Attention Methods

  • Guided Attention: paper
  • Forward Backward Decoding: paper
  • Graves Attention: paper
  • Double Decoder Consistency: blog

Speaker Encoder

Vocoders

You can also help us implement more models. Some TTS related work can be found here.

Install TTS

TTS supports python >= 3.6, <3.9.

If you are only interested in synthesizing speech with the released TTS models, installing from PyPI is the easiest option.

pip install TTS

If you plan to code or train models, clone TTS and install it locally.

git clone https://github.com/mozilla/TTS
pip install -e .

Directory Structure

|- notebooks/       (Jupyter Notebooks for model evaluation, parameter selection and data analysis.)
|- utils/           (common utilities.)
|- TTS
    |- bin/             (folder for all the executables.)
      |- train*.py                  (train your target model.)
      |- distribute.py              (train your TTS model using Multiple GPUs.)
      |- compute_statistics.py      (compute dataset statistics for normalization.)
      |- convert*.py                (convert target torch model to TF.)
    |- tts/             (text to speech models)
        |- layers/          (model layer definitions)
        |- models/          (model definitions)
        |- tf/              (Tensorflow 2 utilities and model implementations)
        |- utils/           (model specific utilities.)
    |- speaker_encoder/ (Speaker Encoder models.)
        |- (same)
    |- vocoder/         (Vocoder models.)
        |- (same)

Sample Model Output

Below you see Tacotron model state after 16K iterations with batch-size 32 with LJSpeech dataset.

"Recent research at Harvard has shown meditating for as little as 8 weeks can actually increase the grey matter in the parts of the brain responsible for emotional regulation and learning."

Audio examples: soundcloud

example_output

Datasets and Data-Loading

TTS provides a generic dataloader easy to use for your custom dataset. You just need to write a simple function to format the dataset. Check datasets/preprocess.py to see some examples. After that, you need to set dataset fields in config.json.

Some of the public datasets that we successfully applied TTS:

Example: Synthesizing Speech on Terminal Using the Released Models.

After the installation, TTS provides a CLI interface for synthesizing speech using pre-trained models. You can either use your own model or the release models under the TTS project.

Listing released TTS models.

tts --list_models

Run a tts and a vocoder model from the released model list. (Simply copy and paste the full model names from the list as arguments for the command below.)

tts --text "Text for TTS" \
    --model_name "<type>/<language>/<dataset>/<model_name>" \
    --vocoder_name "<type>/<language>/<dataset>/<model_name>" \
    --out_path folder/to/save/output/

Run your own TTS model (Using Griffin-Lim Vocoder)

tts --text "Text for TTS" \
    --model_path path/to/model.pth.tar \
    --config_path path/to/config.json \
    --out_path output/path/speech.wav

Run your own TTS and Vocoder models

tts --text "Text for TTS" \
    --model_path path/to/config.json \
    --config_path path/to/model.pth.tar \
    --out_path output/path/speech.wav \
    --vocoder_path path/to/vocoder.pth.tar \
    --vocoder_config_path path/to/vocoder_config.json

Note: You can use ./TTS/bin/synthesize.py if you prefer running tts from the TTS project folder.

Example: Training and Fine-tuning LJ-Speech Dataset

Here you can find a CoLab notebook for a hands-on example, training LJSpeech. Or you can manually follow the guideline below.

To start with, split metadata.csv into train and validation subsets respectively metadata_train.csv and metadata_val.csv. Note that for text-to-speech, validation performance might be misleading since the loss value does not directly measure the voice quality to the human ear and it also does not measure the attention module performance. Therefore, running the model with new sentences and listening to the results is the best way to go.

shuf metadata.csv > metadata_shuf.csv
head -n 12000 metadata_shuf.csv > metadata_train.csv
tail -n 1100 metadata_shuf.csv > metadata_val.csv

To train a new model, you need to define your own config.json to define model details, trainin configuration and more (check the examples). Then call the corressponding train script.

For instance, in order to train a tacotron or tacotron2 model on LJSpeech dataset, follow these steps.

python TTS/bin/train_tacotron.py --config_path TTS/tts/configs/config.json

To fine-tune a model, use --restore_path.

python TTS/bin/train_tacotron.py --config_path TTS/tts/configs/config.json --restore_path /path/to/your/model.pth.tar

To continue an old training run, use --continue_path.

python TTS/bin/train_tacotron.py --continue_path /path/to/your/run_folder/

For multi-GPU training, call distribute.py. It runs any provided train script in multi-GPU setting.

CUDA_VISIBLE_DEVICES="0,1,4" python TTS/bin/distribute.py --script train_tacotron.py --config_path TTS/tts/configs/config.json

Each run creates a new output folder accomodating used config.json, model checkpoints and tensorboard logs.

In case of any error or intercepted execution, if there is no checkpoint yet under the output folder, the whole folder is going to be removed.

You can also enjoy Tensorboard, if you point Tensorboard argument--logdir to the experiment folder.

Contribution Guidelines

This repository is governed by Mozilla's code of conduct and etiquette guidelines. For more details, please read the Mozilla Community Participation Guidelines.

  1. Create a new branch.
  2. Implement your changes.
  3. (if applicable) Add Google Style docstrings.
  4. (if applicable) Implement a test case under tests folder.
  5. (Optional but Prefered) Run tests.
./run_tests.sh
  1. Run the linter.
pip install pylint cardboardlint
cardboardlinter --refspec master
  1. Send a PR to dev branch, explain what the change is about.
  2. Let us discuss until we make it perfect :).
  3. We merge it to the dev branch once things look good.

Feel free to ping us at any step you need help using our communication channels.

Collaborative Experimentation Guide

If you like to use TTS to try a new idea and like to share your experiments with the community, we urge you to use the following guideline for a better collaboration. (If you have an idea for better collaboration, let us know)

  • Create a new branch.
  • Open an issue pointing your branch.
  • Explain your idea and experiment.
  • Share your results regularly. (Tensorboard log files, audio results, visuals etc.)

Major TODOs

Acknowledgement

More Repositories

1

pdf.js

PDF Reader in JavaScript
JavaScript
43,965
star
2

DeepSpeech

DeepSpeech is an open source embedded (offline, on-device) speech-to-text engine which can run in real time on devices ranging from a Raspberry Pi 4 to high power GPU servers.
C++
24,930
star
3

send

Simple, private file sharing from the makers of Firefox
FreeMarker
13,232
star
4

sops

Simple and flexible tool for managing secrets
Go
12,778
star
5

BrowserQuest

A HTML5/JavaScript multiplayer game experiment
JavaScript
9,167
star
6

nunjucks

A powerful templating engine with inheritance, asynchronous control, and more (jinja2 inspired)
JavaScript
8,522
star
7

geckodriver

WebDriver for Firefox
7,047
star
8

readability

A standalone version of the readability lib
JavaScript
6,470
star
9

sccache

Sccache is a ccache-like tool. It is used as a compiler wrapper and avoids compilation when possible. Sccache has the capability to utilize caching in remote storage environments, including various cloud storage options, or alternatively, in local storage.
Rust
5,661
star
10

mozjpeg

Improved JPEG encoder.
C
5,216
star
11

Fira

Mozilla's new typeface, used in Firefox OS
CSS
4,920
star
12

rhino

Rhino is an open-source implementation of JavaScript written entirely in Java
JavaScript
4,112
star
13

shumway

Shumway is a Flash VM and runtime written in JavaScript
TypeScript
3,692
star
14

source-map

Consume and generate source maps.
JavaScript
3,530
star
15

gecko-dev

Read-only Git mirror of the Mercurial gecko repositories at https://hg.mozilla.org. How to contribute: https://firefox-source-docs.mozilla.org/contributing/contribution_quickref.html
2,897
star
16

web-ext

A command line tool to help build, run, and test web extensions
JavaScript
2,663
star
17

multi-account-containers

Firefox Multi-Account Containers lets you keep parts of your online life separated into color-coded tabs that preserve your privacy. Cookies are separated by container, allowing you to use the web with multiple identities or accounts simultaneously.
JavaScript
2,657
star
18

bleach

Bleach is an allowed-list-based HTML sanitizing library that escapes or strips markup and attributes
Python
2,638
star
19

node-convict

Featureful configuration management library for Node.js
JavaScript
2,334
star
20

cbindgen

A project for generating C bindings from Rust code
Rust
2,314
star
21

MozDef

DEPRECATED - MozDef: Mozilla Enterprise Defense Platform
Python
2,166
star
22

popcorn-js

The HTML5 Media Framework. (Unmaintained. See https://github.com/menismu/popcorn-js for activity)
JavaScript
2,148
star
23

webextension-polyfill

A lightweight polyfill library for Promise-based WebExtension APIs in Chrome
JavaScript
2,088
star
24

fathom

A framework for extracting meaning from web pages
JavaScript
1,969
star
25

cipherscan

A very simple way to find out which SSL ciphersuites are supported by a target.
Python
1,912
star
26

hawk

HTTP Holder-Of-Key Authentication Scheme
JavaScript
1,903
star
27

persona

Persona is a secure, distributed, and easy to use identification system.
JavaScript
1,828
star
28

neqo

Neqo, the Mozilla Firefox implementation of QUIC in Rust
Rust
1,820
star
29

http-observatory

Mozilla HTTP Observatory
Python
1,784
star
30

uniffi-rs

a multi-language bindings generator for rust
Rust
1,783
star
31

mentat

UNMAINTAINED A persistent, relational store inspired by Datomic and DataScript.
Rust
1,650
star
32

task.js

Beautiful concurrency for JavaScript
JavaScript
1,635
star
33

hubs

Duck-themed multi-user virtual spaces in WebVR. Built with A-Frame.
JavaScript
1,561
star
34

fx-private-relay

Keep your email safe from hackers and trackers. Make an email alias with 1 click, and keep your address to yourself.
Python
1,457
star
35

pontoon

Mozilla's Localization Platform
Python
1,442
star
36

thimble.mozilla.org

UPDATE: This project is no longer maintained. Please check out Glitch.com instead.
JavaScript
1,423
star
37

kitsune

Platform for Mozilla Support
Python
1,289
star
38

mig

Distributed & real time digital forensics at the speed of the cloud
Go
1,195
star
39

bedrock

Making mozilla.org awesome, one pebble at a time
HTML
1,155
star
40

grcov

Rust tool to collect and aggregate code coverage data for multiple source files
Rust
1,152
star
41

OpenWPM

A web privacy measurement framework
Python
1,150
star
42

policy-templates

Policy Templates for Firefox
1,117
star
43

server-side-tls

Server side TLS Tools
HTML
1,114
star
44

rust-android-gradle

Kotlin
989
star
45

contain-facebook

Facebook Container isolates your Facebook activity from the rest of your web activity in order to prevent Facebook from tracking you outside of the Facebook website via third party cookies.
JavaScript
967
star
46

pdfjs-dist

Generic build of PDF.js library.
JavaScript
952
star
47

narcissus

INACTIVE - http://mzl.la/ghe-archive - The Narcissus meta-circular JavaScript interpreter
JavaScript
901
star
48

openbadges-backpack

Mozilla Open Badges Backpack
JavaScript
861
star
49

addons-server

๐Ÿ•ถ addons.mozilla.org Django app and API ๐ŸŽ‰
Python
833
star
50

awsbox

INACTIVE - http://mzl.la/ghe-archive - A featherweight PaaS on top of Amazon EC2 for deploying node apps
JavaScript
811
star
51

dxr

DEPRECATED - Powerful search for large codebases
Python
804
star
52

ssh_scan

DEPRECATED - A prototype SSH configuration and policy scanner (Blog: https://mozilla.github.io/ssh_scan/)
Ruby
793
star
53

chromeless

DEPRECATED - Build desktop applications with web technologies.
JavaScript
761
star
54

node-client-sessions

secure sessions stored in cookies
JavaScript
745
star
55

playdoh

PROJECT DEPRECATED (WAS: "Mozilla's Web application base template. Half Django, half awesomeness, half not good at math.")
Python
714
star
56

blurts-server

Mozilla Monitor arms you with tools to keep your personal information safe. Find out what hackers already know about you and learn how to stay a step ahead of them.
Fluent
709
star
57

DeepSpeech-examples

Examples of how to use or integrate DeepSpeech
Python
682
star
58

tofino

Project Tofino is a browser interaction experiment.
HTML
655
star
59

addon-sdk

DEPRECATED - The Add-on SDK repository.
641
star
60

cargo-vet

supply-chain security for Rust
Rust
636
star
61

MozStumbler

Android Stumbler for Mozilla
Java
621
star
62

standards-positions

Python
619
star
63

application-services

Firefox Application Services
Rust
605
star
64

lightbeam

Orignal unmaintained version of the Lightbeam extension. See lightbeam-we for the new one which works in modern versions of Firefox.
JavaScript
587
star
65

fxa

Monorepo for Mozilla Accounts (formerly Firefox Accounts)
JavaScript
580
star
66

firefox-translations

Firefox Translations is a webextension that enables client side translations for web browsers.
JavaScript
575
star
67

moz-sql-parser

DEPRECATED - Let's make a SQL parser so we can provide a familiar interface to non-sql datastores!
Python
574
star
68

spidernode

Node.js on top of SpiderMonkey
JavaScript
560
star
69

ichnaea

Mozilla Ichnaea
Python
559
star
70

inclusion

Our repository for Diversity, Equity and Inclusion work at Mozilla
557
star
71

positron

a experimental, Electron-compatible runtime on top of Gecko
551
star
72

addons-frontend

Front-end to complement mozilla/addons-server
JavaScript
525
star
73

tls-observatory

An observatory for TLS configurations, X509 certificates, and more.
Go
518
star
74

nixpkgs-mozilla

Mozilla overlay for Nixpkgs.
Nix
510
star
75

neo

INACTIVE - http://mzl.la/ghe-archive - DEPRECATED: See https://neutrino.js.org for alternative
JavaScript
503
star
76

notes

DEPRECATED - A notepad for Firefox
HTML
495
star
77

bugbug

Platform for Machine Learning projects on Software Engineering
Python
493
star
78

django-csp

Content Security Policy for Django.
Python
486
star
79

skywriter

Mozilla Skywriter
JavaScript
481
star
80

Spoke

Easily create custom 3D environments
JavaScript
480
star
81

zamboni

Backend for the Firefox Marketplace
Python
474
star
82

vtt.js

A JavaScript implementation of the WebVTT specification
JavaScript
461
star
83

FirefoxColor

Theming demo for Firefox Quantum and beyond
JavaScript
452
star
84

libdweb

Extension containing an experimental libdweb APIs
JavaScript
441
star
85

mozilla-django-oidc

A django OpenID Connect library
Python
435
star
86

pointer.js

INACTIVE - http://mzl.la/ghe-archive - INACTIVE - http://mzl.la/ghe-archive - Normalizes mouse/touch events into 'pointer' events.
JavaScript
435
star
87

agithub

Agnostic Github client API -- An EDSL for connecting to REST servers
Python
414
star
88

cubeb

Cross platform audio library
C++
411
star
89

fxa-auth-server

DEPRECATED - Migrated to https://github.com/mozilla/fxa
JavaScript
401
star
90

zilla-slab

Mozilla's Zilla Slab Type Family
Shell
398
star
91

r2d2b2g

Firefox OS Simulator is a test environment for Firefox OS. Use it to test your apps in a Firefox OS-like environment that looks and feels like a mobile phone.
JavaScript
391
star
92

masche

Deprecated - MIG Memory Forensic library
Go
387
star
93

qbrt

CLI to a Gecko desktop app runtime
JavaScript
386
star
94

mp4parse-rust

Parser for ISO Base Media Format aka video/mp4 written in Rust.
Rust
380
star
95

valence

INACTIVE - http://mzl.la/ghe-archive - Firefox Developer Tools protocol adapters (Unmaintained)
JavaScript
377
star
96

OpenDesign

Mozilla Open Design aims to bring open source principles to Creative Design. Find us on Matrix: chat.mozilla.org/#/room/#opendesign:mozilla.org
370
star
97

reflex

Functional reactive UI library
JavaScript
364
star
98

mortar

INACTIVE - http://mzl.la/ghe-archive - A collection of web app templates
364
star
99

ssl-config-generator

Mozilla SSL Configuration Generator
Handlebars
358
star
100

minion

Minion
354
star