kyutai-labs/pocket-tts Alternatives

choose if you need state-of-the-art multilingual TTS and have GPU resources.

VoxCPM2: Tokenizer-Free TTS for Multilingual Speech Generation, Creative Voice Design, and True-to-Life Cloning

choose if you need industrial-level controllable zero-shot TTS and can run large GPU models.

An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

choose if you need a full-stack multilingual TTS with training support and GPU deployment.

Multi-lingual large voice generation model, providing inference, training and deployment full-stack ability.

choose if you need voice cloning for 600+ languages, accepting higher resource use.

High-Quality Voice Cloning TTS for 600+ Languages

choose if you prioritize expressive open-weight TTS and have GPU; note the repo is stale.

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 if you need strong few-shot cloning and can train/run heavy GPU models.

1 min voice data can also be used to train a good TTS model! (few shot voice cloning)

choose if you want so-called SoTA TTS quality with GPU availability.

SoTA open-source TTS

choose if you need a cross-platform C++/embedded inference engine that includes TTS.

Speech-to-text, text-to-speech, speaker diarization, speech enhancement, source separation, and VAD using next-gen Kaldi with onnxruntime without Internet connection. Support embedded systems, Android, iOS, HarmonyOS, Raspberry Pi, RISC-V, RK NPU, Axera NPU, Ascend NPU, x86_64 servers, websocket server/client, support 12 programming languages

choose if you want an all-in-one speech toolkit with TTS/ASR on PaddlePaddle.

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 if you want Spark-TTS's specific model and can run GPU inference.

Spark-TTS Inference Code

choose if you need a low-latency voice-agent stack with TTS/STT on edge devices.

Very low latency speech to text, intent recognition, and text to speech, for building voice agents and interfaces

choose if you're on Apple Silicon and want MLX-accelerated TTS.

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.

choose if you want cloud TTS without local model overhead and have internet connectivity.

Use Microsoft Edge's online text-to-speech service from Python WITHOUT needing Microsoft Edge or Windows or an API key

choose if you need prompt-controlled emotion in TTS and have GPU.

EmotiVoice 😊: a Multi-Voice and Prompt-Controlled TTS Engine

choose if you want a TTS built by inverting Whisper, and have GPU.

An Open Source text-to-speech system built by inverting Whisper.

choose if you want a human-sounding expressive TTS and can run GPU.

Towards Human-Sounding Speech

choose if you want an expressive foundational TTS and have GPU.

Foundational model for human-like, expressive TTS

choose if you need a mature TTS library with many models and accept heavier compute and no recent maintenance.

🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production

choose if you need a proven local CPU TTS with many prebuilt voices but lack voice cloning.

A fast, local neural text to speech system

choose if you need a lightweight multi-lingual TTS and accept an unmaintained project.

High-quality multi-lingual text-to-speech library by MyShell.ai. Support English, Spanish, French, Chinese, Japanese and Korean.

choose if you want a tiny, realtime multilingual TTS with Apache-2.0 but no voice cloning.

MOSS-TTS-Nano is an open-source multilingual tiny speech generation model from MOSI.AI and the OpenMOSS team. With only 0.1B parameters, it is designed for realtime speech generation, can run directly on CPU without a GPU, and keeps the deployment stack simple enough for local demos, web serving, and lightweight product integration.