F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching
Paper • 2410.06885 • Published • 48
F5-TTS Hindi 24KHz Model
This is a Hindi Text-to-Speech model trained from scratch using the F5 architecture.
Details
- Developed by: SPRING Lab, Indian Institute of Technology, Madras
- Language: Hindi
- License: CC-BY-4.0
Uses
The model was developed and is primarily intended for research purposes.
How to Get Started with the Model
Clone the following github repo and refer to the README: https://github.com/rumourscape/F5-TTS
Training Details
The model was trained on 8x A100 40GB GPUs for close to a week. We would like to thank CDAC for providing the compute resources.
We used the "small" configuration(151M parameter) model for training according to the F5 paper.
Training Data
We used the Hindi subsets of IndicTTS and IndicVoices-R datasets for training this model.
- IndicTTS-Hindi: https://huggingface.co/datasets/SPRINGLab/IndicTTS-Hindi
- IndicVoices-R_Hindi: https://huggingface.co/datasets/SPRINGLab/IndicVoices-R_Hindi
- Downloads last month
- 857
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js