All models
dialogue

Dia

Apache-2.0

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

Dia generates full conversations directly from screenplay-style text — tag lines with [S1] and [S2] and it renders a natural two-voice dialogue in a single pass, complete with laughs, coughs, and throat-clears. A 1.6B model that went viral for podcast-style output rivaling paid tools.

Voices

2 voices
Female
Speaker 1default
american
Male
Speaker 2
american

The scripts

Neutral
“The quick brown fox jumps over the lazy dog near the riverbank.”
Emotional
“I can't believe you actually did it. This is incredible!”
Numbers & Dates
“On March 14th, 2025, the team raised $4.2 million at a 38% margin.”

Install

pip install nari-tts

Dia FAQ

Do I need a GPU to run Dia?

For practical speeds, yes — plan on roughly 10 GB of VRAM. If you'd rather not manage a GPU, hosted access to every model in this directory is coming via OpenSpeech Cloud.

Can Dia clone voices?

Yes — Dia supports voice cloning from reference audio.

What license is Dia released under?

Dia is released under the Apache-2.0 license. Always check the repository for the exact terms — some models license code and weights separately.

What languages does Dia support?

Dia supports English.

Is there a hosted Dia API?

Yes — OpenSpeech Cloud is launching hosted, pay-per-minute access to Dia and every other model in this directory. No CUDA, Docker, or GPU needed.