Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
TTS model capable of streaming conversational audio in realtime.
| Date | Stars |
|---|---|
| 2026-07-24 | 1160 |
| 2026-07-25 | 1160 |
| 2026-07-28 | 1160 |
| 2026-07-30 | 1160 |
| 2026-08-06 | 1160 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<div align="center">
<a href="https://huggingface.co/nari-labs/Dia2-2B"><img src="https://img.shields.io/badge/HF%20Repo-Dia2--2B-orange?style=for-the-badge"></a>
<a href="https://discord.gg/bJq6vjRRKv"><img src="https://img.shields.io/badge/Discord-Join%20Chat-7289DA?logo=discord&style=for-the-badge"></a>
<a href="https://github.com/nari-labs/dia2/blob/main/LICENSE"><img src="https://img.shields.io/badge/License-Apache_2.0-blue.svg?style=for-the-badge"></a>
</div>
**Dia2** is a **streaming dialogue TTS model** created by Nari Labs.
The model does not need the entire text to produce the audio, and can start generating as the first few words are given as input. You can condition the output on audio, enabling natural conversations in realtime.
We provide model checkpoints (1B, 2B) and inference code to accelerate research. The model only supports up to 2 minutes of generation in English.
⚠️ Quality and voices vary per generation, as the model is not fine-tuned on a specific voice. Use with prefix or fine-tune in order to obtain stable output.
Try it now on Hugging Face [Spaces](https://huggingface.co/spaces/nari-labs/Dia2-2B)
## Upcoming
- Bonsai (JAX) implementation
- Dia2 TTS Server: Real streaming support
- Sori: Dia2-powered speech-to-speech engine written in Rust
## Quickstart
> **Requirement** — install [uv](https://docs.astral.sh/uv/) and use CUDA 12.8+
> drivers. All commands below run through `uv run …` as a rule.
1. **Install dependencies (one-time):**
```bash
uv sync
```
2. **Prepare a script:** edit `input.txt` using `[S1]` / `[S2]` speaker tags.
3. **Generate audio:**
```bash
uv run -m dia2.cli \
--hf nari-labs/Dia2-2B \
--input input.txt \
--cfg 6.0 --temperature 0.8 \
--cuda-graph --verbose \
output.wav
```
The first run downloads weights/tokenizer/Mimi. The CLI auto-selects CUDA when available (otherwise CPU) and defaults to bfloat16 precision—override with `--device` / `--dtype` if needed.
4. **Conditional Generation (recommended for stable use):**
```bash
uv run -m dia2.cli \
--hf nari-labs/Dia2-2B \
--input input.txt \
--prefix-speaker-1 example_prefix1.wav \
--prefix-speaker-2 example_prefix2.wav \
--cuda-graph --verbose \
output_conditioned.wav
```
Condition the generation on previous conversational context in order to generate natural output for your speech-to-speech system. For example, place the voice of your assistant as prefix speaker 1, place user's audio input as prefix speaker 2, and generate the response to user's input.
Whisper is used to transcribe each prefix file, which takes additional time. We include example prefix files as `example_prefix1.wav` and `example_prefix2.wav` (both files are output created by the model).
6. **Gradio for Easy Usage**
```bash
uv run gradio_app.py
```
### Programmatic Usage
```python
from dia2 import Dia2, GenerationConfig, SamplingConfig
dia = Dia2.from_repo("nari-labs/Dia2-2B", device="cuda", dtype="bfloat16")
config = GenerationConfig(
cfg_scale=2.0,
audio=SamplingConfig(temperature=0.8, top_k=50),
use_cuda_graph=True,
)
result = dia.generate("[S1] Hello Dia2!", config=config, output_wav="hello.wav", verbose=True)
```
Generation runs until the runtime config's `max_context_steps` (1500, 2 minutes)
or until EOS is detected. `GenerationResult` includes audio tokens, waveform tensor,
and word timestamps relative to Mimi’s ~12.5 Hz frame rate.
## Hugging Face
| Variant | Repo |
| --- | --- |
| Dia2-1B | [`nari-labs/Dia2-1B`](https://huggingface.co/nari-labs/Dia2-1B)
| Dia2-2B | [`nari-labs/Dia2-2B`](https://huggingface.co/nari-labs/Dia2-2B)
## License & Attribution
Licensed under [Apache 2.0](LICENSE). All third-party assets (Kyutai Mimi codec, etc.) retain their original licenses.
## Disclaimer
This project offers a high-fidelity speech generation model intended for research and educational use. The followingExcerpt of 4,918 characters
Read on GitHub
Toby Kim · Something New Labs
12
Jaeyong Sung · Cartesia · South Korea
4
Jakub Czarlinski
2
Peter Evers · Mistral AI · Netherlands
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:6087a1aa38d79d9f, topic:text-to-speech