F5-TTS

This is a mirror of the original weights for use with TTSDB.

Original weights: https://huggingface.co/SWivid/F5-TTS Original code: https://github.com/SWivid/F5-TTS

Non-Autoregressive Flow Matching (DiT) text-to-speech model by Yushen Chen.

Original Work

This model was created by the original authors. Please cite their work if you use this model:

@inproceedings{f5-tts,
  title = "F5-{TTS}: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching",
  author = "Chen, Yushen  and
    Niu, Zhikang  and
    Ma, Ziyang  and
    Deng, Keqi  and
    Wang, Chunhui  and
    JianZhao, JianZhao  and
    Yu, Kai  and
    Chen, Xie",
  editor = "Che, Wanxiang  and
    Nabende, Joyce  and
    Shutova, Ekaterina  and
    Pilehvar, Mohammad Taher",
  booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
  month = jul,
  year = "2025",
  address = "Vienna, Austria",
  publisher = "Association for Computational Linguistics",
  url = "https://aclanthology.org/2025.acl-long.313/",
  doi = "10.18653/v1/2025.acl-long.313",
  pages = "6255--6271",
  ISBN = "979-8-89176-251-0",
  abstract = "This paper introduces F5-TTS, a fully non-autoregressive text-to-speech system based on flow matching with Diffusion Transformer (DiT). Without requiring complex designs such as duration model, text encoder, and phoneme alignment, the text input is simply padded with filler tokens to the same length as input speech, and then the denoising is performed for speech generation, which was originally proved feasible by E2 TTS. However, the original design of E2 TTS makes it hard to follow due to its slow convergence and low robustness. To address these issues, we first model the input with ConvNeXt to refine the text representation, making it easy to align with the speech. We further propose an inference-time Sway Sampling strategy, which significantly improves our model{'}s performance and efficiency. This sampling strategy for flow step can be easily applied to existing flow matching based models without retraining. Our design allows faster training and achieves an inference RTF of 0.15, which is greatly improved compared to state-of-the-art diffusion-based TTS models. Trained on a public 100K hours multilingual dataset, our F5-TTS exhibits highly natural and expressive zero-shot ability, seamless code-switching capability, and speed control efficiency. We have released all codes and checkpoints to promote community development, at https://SWivid.github.io/F5-TTS/."
}

Papers:

https://aclanthology.org/2025.acl-long.313/

Installation

pip install ttsdb-f5-tts

Usage

from ttsdb_f5_tts import F5TTS

# Load the model (downloads weights automatically)
model = F5TTS(model_id="ttsds/f5-tts")

# Synthesize speech
audio, sample_rate = model.synthesize(
    text="Hello, this is a test of F5-TTS.",
    reference_audio="path/to/reference.wav",
)

# Save the output
model.save_audio(audio, sample_rate, "output.wav")

Model Details

Property	Value
Sample Rate	24000 Hz
Parameters	335M
Architecture	Non-Autoregressive, Flow Matching, Diffusion Transformer
Languages	English, Chinese
Release Date	2024-10-30

Training Data

Emilia Dataset (100000 hours)

License

Weights: Creative Commons Attribution-NonCommercial 4.0
Code: MIT License

Please refer to the original repositories for full license terms.

Model tree for ttsds/f5-tts

Base model

SWivid/F5-TTS

Finetuned

(74)

this model

ttsds
/

f5-tts