text-to-speech / speech synthesis
Coqui TTS is a deep learning toolkit for text-to-speech that provides pretrained models, training and fine-tuning tools, and voice cloning capabilities.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
A general speech toolkit with TTS capabilities; choose it when you work in PaddlePaddle or need ASR and speaker tasks alongside TTS in one framework.
Easy-to-use Speech Toolkit including Self-Supervised Learning model, SOTA/Streaming ASR with punctuation, Streaming TTS with text frontend, Speaker Verification System, End-to-End Speech Translation and Keyword Spotting. Won NAACL2022 Best Demo Award.
An end-to-end speech processing toolkit; choose it if you want a comprehensive PyTorch framework covering ASR and TTS.
End-to-End Speech Processing Toolkit
An audio, music, and speech generation toolkit; choose it for broad research flexibility beyond just TTS.
Amphion (/æmˈfaɪən/) is a toolkit for Audio, Music, and Speech Generation. Its purpose is to support reproducible research and help junior researchers and engineers get started in the field of audio, music, and speech generation research and development.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
A standalone real-time voice-cloning TTS; choose it when you only need cloning and prefer a simpler, separate PyTorch system over Coqui's full toolkit.
🚀Clone a voice in 5 seconds to generate arbitrary speech in real-time
An expressive TTS foundation model; pick it for human-like prosody when you do not need Coqui's extensive multi-model toolkit.
Foundational model for human-like, expressive TTS
A prompt-controlled multi-voice TTS engine; pick it for emotion/prompt-based synthesis instead of Coqui's standard multi-speaker models.
EmotiVoice 😊: a Multi-Voice and Prompt-Controlled TTS Engine
A real-time voice cloning system; choose it for a ready-to-use cloning tool with a notebook interface.
Clone a voice in 5 seconds to generate arbitrary speech in real-time
A fast, local C++ TTS inference engine; pick it for offline edge deployment where Python and Coqui's larger runtime are not suitable.
A fast, local neural text to speech system
A multilingual tokenizer-free TTS model with voice cloning; choose it for creative voice design and true-to-life cloning without Coqui's toolkit.
VoxCPM2: Tokenizer-Free TTS for Multilingual Speech Generation, Creative Voice Design, and True-to-Life Cloning
A zero-shot multi-speaker TTS and voice conversion model; pick it for a single-model zero-shot approach.
YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyone
A few-shot voice cloning TTS that trains on very little data; choose it when you need rapid adaptation with 1 minute of audio.
1 min voice data can also be used to train a good TTS model! (few shot voice cloning)
The reference implementation of the VITS model; pick it for a clean single-model baseline rather than a full library.
VITS: Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech
A high-quality voice cloning TTS for 600+ languages; pick it for outstanding language coverage and production cloning.
High-Quality Voice Cloning TTS for 600+ Languages
A TTS implementation using denoising diffusion GANs; pick for research into diffusion-based speech generation.
PyTorch Implementation of DiffGAN-TTS: High-Fidelity and Efficient Text-to-Speech with Denoising Diffusion GANs
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
The direct predecessor of Coqui TTS; choose it only if you specifically need the historical Mozilla codebase.
:robot: :speech_balloon: Deep learning for Text to Speech (Discussion forum: https://discourse.mozilla.org/c/tts)
An older TTS toolkit for TensorFlow 2; pick it only if you are committed to TF and accept stale code.
:stuck_out_tongue_closed_eyes: TensorFlowTTS: Real-Time State-of-the-art Speech Synthesis for Tensorflow 2 (supported including English, French, Korean, Chinese, German and Easy to adapt for other languages)
An older PaddlePaddle TTS toolkit; useful only for legacy PaddlePaddle users.
PAddle PARAllel text-to-speech toolKIT (supporting Tacotron2, Transformer TTS, FastSpeech2/FastPitch, SpeedySpeech, WaveFlow and Parallel WaveGAN)
These projects were analysed and named TTS among their alternatives. The relationship is not symmetric — how TTS rates them is a separate judgement, made when TTS is analysed in its own right.
calls TTS “Choose it when you need a production-grade TTS toolkit with many models and APIs, trading research simplicity for flexibility.”
Clone a voice in 5 seconds to generate arbitrary speech in real-time
calls TTS “Choose for a multipurpose TTS toolkit with several model options and cloning support, though development has stalled.”
🚀Clone a voice in 5 seconds to generate arbitrary speech in real-time
calls TTS “For a full TTS toolkit with training abilities, if you need fine-tuning instead of just inference.”
Lightning-Fast, On-Device, Multilingual TTS — running natively via ONNX.
calls TTS “Choose when you want a mature, battle-tested local TTS toolkit with many model choices; stale but works.”
Use Microsoft Edge's online text-to-speech service from Python WITHOUT needing Microsoft Edge or Windows or an API key
calls TTS “Choose when you need a production-tested TTS toolkit with many models, though it does not cover music/audio generation like Amphion.”
Amphion (/æmˈfaɪən/) is a toolkit for Audio, Music, and Speech Generation. Its purpose is to support reproducible research and help junior researchers and engineers get started in the field of audio, music, and speech generation research and development.
calls TTS “Choose for a mature Python TTS toolkit with many models and training support.”
Desktop application for neural speech synthesis written in C++
calls TTS “Choose if you need a multi-model TTS training toolkit; it is unmaintained and lacks VoxCPM's modern output quality.”
VoxCPM2: Tokenizer-Free TTS for Multilingual Speech Generation, Creative Voice Design, and True-to-Life Cloning
calls TTS “Older TTS toolkit with no updates in two years; not suitable for new projects.”
SoTA open-source TTS
calls TTS “Mature TTS-only library, unmaintained for 2 years; use if you need a TTS-only solution and can maintain it yourself.”
Easy-to-use Speech Toolkit including Self-Supervised Learning model, SOTA/Streaming ASR with punctuation, Streaming TTS with text frontend, Speaker Verification System, End-to-End Speech Translation and Keyword Spotting. Won NAACL2022 Best Demo Award.
calls TTS “Choose when you only need TTS and want a mature, battle-tested library, even though it's not as broad or recently updated as ESPnet.”
End-to-End Speech Processing Toolkit
calls TTS “Use for a battle-tested TTS toolkit, but unmaintained for over 2 years.”
High-Quality Voice Cloning TTS for 600+ Languages
calls TTS “choose if you need a mature TTS library with many models and accept heavier compute and no recent maintenance.”
A TTS that fits in your CPU (and pocket)
calls TTS “choose as a research toolkit to train custom models, but development has stalled.”
EmotiVoice 😊: a Multi-Voice and Prompt-Controlled TTS Engine
calls TTS “Choose when you need a mature PyTorch TTS toolkit with many voices and can accept an unmaintained codebase.”
A text-to-speech (TTS), speech-to-text (STT) and speech-to-speech (STS) library built on Apple's MLX framework, providing efficient speech analysis on Apple Silicon.
calls TTS “A mature TTS toolkit with a large model zoo, but unmaintained for two years; choose it if you need many models and can accept stale releases.”
High-quality multi-lingual text-to-speech library by MyShell.ai. Support English, Spanish, French, Chinese, Japanese and Korean.