Bark
Suno team's open-source TTS (before Suno did music) — famed for realistic interjections
The Suno team's open-source TTS (from before Suno made music); went viral for its realistic interjections.
Text-to-speech, speech-to-text, and voice cloning, open source.
Key features
- Transformer text-to-audio model for speech, music, sound effects
- Multilingual speech with automatic language detection
- Nonverbal sounds: laughing, sighing, crying
- 100+ speaker presets; MIT licensed for commercial use
Good for
- Creative audio generation beyond conventional TTS
- Multilingual narration with expressive voice presets
Stack & tags
Pythontext-to-audiottsmultilingualsound-effectsmusic-generationopen-weightssuno
Screenshots
Screenshots from the project’s GitHub README.
Similar open-source projects
Whisper
★ 109,795OpenAI's open-source speech recognition model — the benchmark for multilingual STT
GPT-SoVITS
★ 62,271TTS/cloning trainable on 1 minute of audio — a legendary Chinese-community project