text-to-speech (TTS) and voice cloning
Chatterbox is a family of state-of-the-art open-source multilingual text-to-speech models with voice cloning and expressive speech generation.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
Choose for tokenizer-free multilingual TTS with active updates and creative voice design.
VoxCPM2: Tokenizer-Free TTS for Multilingual Speech Generation, Creative Voice Design, and True-to-Life Cloning
Choose for industrial-grade zero-shot TTS with controllable and efficient inference.
An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System
Choose for a full-stack multilingual TTS with training and deployment support.
Multi-lingual large voice generation model, providing inference, training and deployment full-stack ability.
Choose for high-quality voice-cloning TTS supporting 600+ languages.
High-Quality Voice Cloning TTS for 600+ Languages
Choose for an established open-weight TTS with high multilingual expressiveness.
Zonos-v0.1 is a leading open-weight text-to-speech model trained on more than 200k hours of varied multilingual speech, delivering expressiveness and quality on par with—or even surpassing—top TTS providers.
Choose for few-shot voice cloning with your own data and an active community.
1 min voice data can also be used to train a good TTS model! (few shot voice cloning)
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
Different voice-cloning approach; choose for a simpler, lighter cloning pipeline.
Instant voice cloning by MIT and MyShell. Audio foundation model.
Choose for extremely compact CPU TTS where model size and latency matter more than expressiveness.
A TTS that fits in your CPU (and pocket)
Choose when you need an all-in-one speech toolkit including TTS plus ASR, diarization, etc.
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.
Choose for zero local compute via Microsoft's online TTS service.
Use Microsoft Edge's online text-to-speech service from Python WITHOUT needing Microsoft Edge or Windows or an API key
Choose for a research toolkit covering many audio generation methods, not 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.
Choose for a comprehensive speech processing toolkit with TTS recipes.
End-to-End Speech Processing Toolkit
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
Narrower bilingual TTS with stale code; only if you need its specific voice capabilities.
Spark-TTS Inference Code
Older TTS toolkit with no updates in two years; not suitable for new projects.
🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production
Choose for very fast offline TTS on edge devices, but archived and stale.
A fast, local neural text to speech system
Older emotion-controllable TTS; not updated for years.
EmotiVoice 😊: a Multi-Voice and Prompt-Controlled TTS Engine
Older multilingual TTS library with limited maintenance.
High-quality multi-lingual text-to-speech library by MyShell.ai. Support English, Spanish, French, Chinese, Japanese and Korean.
Previous-generation voice-cloning TTS with narrower scope and older architecture.
Clone a voice in 5 seconds to generate arbitrary speech in real-time
Does the same job, but ships as an app. Useful to a person, not swappable into a codebase.
End-user AI voice studio, not a code-swappable TTS model.
The open-source AI voice studio. Clone, dictate, create.
These projects were analysed and named chatterbox among their alternatives. The relationship is not symmetric — how chatterbox rates them is a separate judgement, made when chatterbox is analysed in its own right.
calls chatterbox “choose over Zonos for a newer MIT-licensed SoTA open TTS from Resemble AI, especially if you value that ecosystem or want a second high-quality baseline.”
Zonos-v0.1 is a leading open-weight text-to-speech model trained on more than 200k hours of varied multilingual speech, delivering expressiveness and quality on par with—or even surpassing—top TTS providers.
calls chatterbox “For a state-of-the-art TTS model with speech generation and control, if you can handle heavier compute.”
Lightning-Fast, On-Device, Multilingual TTS — running natively via ONNX.
calls chatterbox “Choose when you need SoTA local open TTS with naturalness; requires GPU and model download.”
Use Microsoft Edge's online text-to-speech service from Python WITHOUT needing Microsoft Edge or Windows or an API key
calls chatterbox “choose if you want so-called SoTA TTS quality with GPU availability.”
A TTS that fits in your CPU (and pocket)
calls chatterbox “choose for cloning with emotional expressiveness, but it lacks EmotiVoice's prompt system.”
EmotiVoice 😊: a Multi-Voice and Prompt-Controlled TTS Engine
calls chatterbox “State-of-the-art open-source TTS with expressive output; pick it if you prioritize latest quality and can handle higher compute.”
High-quality multi-lingual text-to-speech library by MyShell.ai. Support English, Spanish, French, Chinese, Japanese and Korean.
calls chatterbox “Choose for SoTA open-source TTS with a permissive MIT license if you want a straightforward TTS swap.”
A powerful 3B-parameter, LLM-based Reinforcement Learning audio edit model excels at editing emotion, speaking style, and paralinguistics, and features robust zero-shot text-to-speech
calls chatterbox “Choose when you want a PyTorch TTS with strong expressiveness rather than MLX-specific inference.”
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.