The full directory.

Generates laughter, sighs, and music — the chaotic one

ElevenLabs-tier quality, MIT licensed, voice cloning from a clip

Alibaba's real-time TTS, 150ms streaming latency

Ultra-realistic two-speaker dialogue, sounds like a podcast

Flow-matching TTS, fast zero-shot voice cloning

Open-source TTS that benchmarks above closed-source

The community's favorite few-shot voice-cloning toolkit

Built on Llama 3.2 3B, 10M+ hours of training audio

Bilibili's TTS with multi-speaker and precise emotion control

15M params, under 25MB, runs anywhere — even your phone

82M params, Apache 2.0, runs on a potato

ByteDance's TTS with sparse alignment — robust prosody

Multilingual, real-time on CPU

On-device-class TTS with instant cloning from ~3 seconds

Tone-color cloning + cross-lingual transfer

Llama-3 based, empathetic, low-latency for interactive apps

Describe the voice in natural language ('soft female, fast, clear')

Tiny, fast, runs offline on a Raspberry Pi

Alibaba's flagship, 97ms latency, 10 languages

The conversational speech model behind the viral Maya demo

Built on Qwen2.5, zero-shot voice cloning + style control

Diffusion-based, human-level naturalness on LibriTTS

Microsoft's long-form TTS — 90 minutes, 4 speakers

Whisper, but inverted — TTS by 'unwrapping' OpenAI's ASR

Most-downloaded TTS on HF, 6-second voice cloning

44kHz studio-quality output with emotion control