multi-speaker-tts 开源项目推荐（19 个）

coqui-ai /TTS

一个用于文本转语音的深度学习工具包，已在研究和生产环境中经过实战检验

remsky /Kokoro-FastAPI

用于 Kokoro-82M 文本转语音的 Docker 化 OpenAI 兼容封装，基于 PyTorch 支持多平台 CPU、AMD、NVIDIA GPU；支持多说话人、克隆调优、字幕时间戳、SSML、可选的跟读 Web UI

TensorSpeech /TensorFlowTTS

TensorFlowTTS：面向 Tensorflow 2 的实时最先进语音合成（支持英语、法语、韩语、中文、德语，且易于适配其他语言）

PaddlePaddle /Parakeet

PAddle并行文本语音转换工具包（支持Tacotron 2，Transformer TTS，FastSpeech 2/FastPitch，SpeedySpeech，WaveFlow和并行WaveGAN）

keonlee9420 /DiffGAN-TTS

PyTorch Implementation of DiffGAN-TTS: High-Fidelity and Efficient Text-to-Speech with Denoising Diffusion GANs

nipponjo /tts-arabic-pytorch

🎙️ Arabic TTS models (Tacotron2, FastPitch)

ga642381 /FastSpeech2

Multi-Speaker Pytorch FastSpeech2: Fast and High-Quality End-to-End Text to Speech ✊

gooofy /zerovox

zero-shot realtime TTS system, fully offline, free and open source

nipponjo /tts_arabic

🎙️ Arabic TTS models (FastPitch, Mixer-TTS) in the ONNX format — Python package for offline speech synthesis 🚀📦

hcy71o /SC-CNN

SC-CNN: Effective Speaker Conditioning Method for Zero-Shot Multi-Speaker Text-to-Speech Systems

msalhab96 /MultiSpeech

pytorch implementation for MultiSpeech: Multi-Speaker Text to Speech with Transformer paper

ssmlkl /MnTTS2

This is the experimental description of MnTTS2.

MahtaFetrat /Persian-MultiSpeaker-Tacotron2

Implementation of Transfer Learning from Speaker Verification to Multi-speaker Text-To-Speech Synthesis (SV2TTS) in Persian language.

nh9k /Custom-Speech-Synthesis-Service-using-Multi-speaker-tacotron

Custom-Speech-Synthesis-Service-using-Multi-speaker-tacotron

starryark /tricolour-tts

Tricolour Lovestory (三色绘恋) character voice TTS: multi-speaker Qwen3-TTS fine-tuning with emotion control for 14 Mandarin game characters. Data pipeline, fixed trainer, pre-registered evaluation.

starryark /bangdream-tts

BanG Dream! character voice TTS: 49 voices in Japanese, Mandarin and English. Qwen3-TTS multi-speaker fine-tuning with language ids, native replay, in-voice pseudo-Mandarin and per-voice RVC; pre-registered evaluation. Unofficial fan project.

xiaguanxi /bajiao

芭蕉 Bajiao：LLM + TTS 驱动的 Android AI 有声书制作与听书应用，支持小说 TXT / EPUB 导入、角色识别、多角色配音、声音克隆与音色设计。AI audiobook creator & player with MiMo text-to-speech.

Adibian /Persian-GST-Tacotron

Training GST-Tacotron for Persian language as a multi-speaker Persian speech synthesis.

y2k3073 /Audio8_TTS

Generate multilingual speech and clone voices with this 0.6B-parameter zero-shot text-to-speech model.