You can deploy this model and start earning money today!

CoRal-project/
roest-v3-chatterbox-350m

chatterbox

text-to-speech

alexandrainst/ftspeech

CoRal-project/coral-tts

alexandrainst/nst-da

alexandrainst/nota

ResembleAI/chatterbox-turbo

finetune

openrail

Views

No views yet

Røst-v3-chatterbox-350m

This is a Danish open sourced state-of-the-art Text to Speech (TTS) model. It is trained as part of the CoRal project by the Alexandra Institute.

Looking to improve this model or adapt it to your domain? The finetuning framework used to train this model is fully open-sourced at coral_chatterbox.

Examples

Settings	Mic	Nic
Dette er en tekst-til-tale model, trænet til at oplæse dansk. Lyt med og bedøm selv, hvordan den håndterer sprogets mange forskellige lyde og rytmer.
temp=0.7, top_p=0.95, top_k=600
temp=0.8, top_p=0.95, top_k=1000
København er Danmarks hovedstad og ligger på øerne Sjælland og Amager, hvor mange turister besøger de smukke kanaler og historiske bygninger.
temp=0.7, top_p=0.95, top_k=600
temp=0.8, top_p=0.95, top_k=1000

Inference Quickstart

Start by installing the required libraries:

$ pip install chatterbox-tts huggingface_hub torchaudio

To just run a simple inference

python

import torchaudio as ta
from chatterbox.tts_turbo import ChatterboxTurboTTS
from huggingface_hub import snapshot_download
import os

REPO_ID = "CoRal-project/roest-v3-chatterbox-350m"
device = "cuda" # Change to cpu if no GPU is available

# Load the model
model_dir = snapshot_download(
        repo_id=REPO_ID,
        token=os.getenv("HF_TOKEN") or True,
        # Optional: Filter to download only what you need
        allow_patterns=["*.safetensors", "*.json", "*.txt", "*.pt", "*.model"],
    )

model = ChatterboxTurboTTS.from_local(model_dir, device)


text = "Hej, hvordan går det? Jeg er dansk Chatterbox, og taler naturligt dansk."
wav = model.generate(text)
ta.save("test.wav", wav, model.sr)

# If you want to synthesize with a different voice, specify the audio prompt
AUDIO_PROMPT_PATH = "YOUR_FILE.wav"
wav = model.generate(text, audio_prompt_path=AUDIO_PROMPT_PATH)
ta.save("test-2.wav", wav, model.sr)

Model Details

The model is a finetuned variant of Chatterbox-Turbo, one of the leading open-source text-to-speech models. The base Chatterbox-Turbo model uses the GPT-2 Medium architecture and tokenizer, and is trained on more than 500,000 hours of high-quality English speech.

We finetuned the model on over 2000 hours of Danish speech. The model supports zero-shot voice cloning on as little as a 10 seconds audio prompt. It is compatible with the original Chatterbox library, thus being easy to setup and retaining the built-in Watermarked outputs.

The model works very well with the two predefined speakers from the CoRal-tts dataset, Mic and Nic.

Usage Recommendations

We have found the model to work best with the following settings:

Parameter	Value
temp	0.7-0.8
top_p	0.95
top_k	600-1000

Generally, lower temperature and lower top_k gives longer breaks and more stable prosody.

Known Limitations

The model does not handle longer text inputs natively. Others have implemented streaming for this use case, and we recommend splitting longer texts into sentences.

Evaluation

The model was evaluated using Mean Opinion Score (MOS), achieving a score of 4.01 as rated by a panel of 20 native Danish speakers.

The evaluation used a set of 10 samples for two different speakers, Mic and Nic. Samples were generated with temp=0.7, top_p=0.95, top_k=600. The MOS scale was defined as follows:

Score	Rating	Description
1.0	Bad	Speech sounds completely unnatural and artificial. The synthetic voice is very obvious and so distracting that it can be hard to listen to.
2.0	Poor	Speech sounds mostly unnatural. The synthetic quality is obvious and feels distracting, but still bearable.
3.0	Fair	Speech appears both natural and unnatural in roughly equal measure. It's clear the voice is synthetic, and it can be somewhat distracting.
4.0	Good	Speech sounds mostly natural. You can tell it's synthetic speech, but it's only slightly noticeable and not distracting.
5.0	Excellent	Speech sounds completely natural and cannot be distinguished from a real human voice. There are no audible signs of synthetic speech.

Training Notes

Hyperparameters

Parameter	Value
Epochs	3
Batch size (per device)	4
Gradient accumulation steps	4
Learning rate	1.0e-4
Warmup steps	500
Precision	FP16
Gradient checkpointing	Disabled
Max gradient norm	2
Seed	42
Eval split size	0.1%

Training Data

The model was trained on the following datasets:

Dataset
CoRal-project/coral-tts
alexandrainst/ftspeech
alexandrainst/nst-da
alexandrainst/nota

Creators and Funders

This model has been trained and the model card written by Daniel Christopher Biørrith at Alexandra Institute.

The CoRal project is funded by the Danish Innovation Fund and consists of the following partners:

Citation

bibtex

@misc{roest-v3-chatterbox-350m,
  author    = {Daniel Christopher Biørrith, Dan Saattrup Nielsen, Sif Bernstorff Lehmann, Simon Leminen Madsen and Torben Blach},
  title     = {Røst-v3-chatterbox-350m: A Danish state-of-the-art text-to-speech model},
  year      = {2026},
  url       = {https://huggingface.co/CoRal-project/roest-v3-chatterbox-350m},
}