AI · AI model
Dia
Open-source 1.6B text-to-speech model generating multi-speaker dialogue in one pass.
Dia is an open-source 1.6-billion parameter voice model developed by Nari Labs that creates conversational audio directly from text transcripts. It supports multi-speaker alternating dialogue alongside nonverbal cues like laughter, sighs, and throat clearing. The model can be conditioned on audio prompts for voice cloning and emotion control, running locally via PyTorch/Transformers or in Hugging Face Spaces.
What it's for
- Generate realistic two-person podcast or conversational dialogue from a transcript
- Add expressive nonverbal sounds like laughter or coughing to synthetic voice clips
- Clone speaker voices and match emotional tone using reference audio prompts
- Run local, private multi-speaker speech synthesis without cloud API fees
Features
- Multi-speaker generation in a single pass using speaker tags ([S1], [S2])
- Nonverbal audio synthesis including laughter, sighs, and throat clearing
- Reference audio prompt conditioning for voice cloning and tone control
- Hugging Face Transformers support alongside native inference scripts
- Built-in Gradio web interface and CLI runner
Pricing
Open source. Free and open source under the Apache-2.0 license.
Details
- Type
- AI model
- Category
- AI
- Topic
- Text to Speech
- Open source
- Yes
- Platforms
- Linux, Windows, macOS, Code library, Command line, Web
- For
- Developers, AI researchers, and creators looking for self-hosted multi-speaker dialogue TTS.
- Alternative to
- ElevenLabs
- Tags
- tts, text to speech, voice cloning, ai audio, open weights, speech synthesis
Why it was saved
- @nick_saraev: Provides a free, unlimited, open-weights alternative to ElevenLabs capable of handling conversational scripts and nonverbal sounds like laughter and sighs.
- @victor__martynov: Serves as an open-source alternative to ElevenLabs and NotebookLM Audio Overviews, allowing precise script-level control over natural emotional delivery, voice switching, laughter, and pauses.
Compare
Related
AI voice generator, voice cloning, and conversational agent platform
Ultra-low-latency streaming text-to-speech and instant voice cloning model.
Fast, low-cost text-to-speech API and studio for creators and developers
Converts spoken thoughts into structured, polished writing in customizable styles.
Official Python library for Cartesia's real-time text-to-speech API
Open-source zero-shot text-to-speech and voice cloning model
Updated