F5-TTS/src/f5_tts/infer/SHARED.md

Shared Model Cards

Prerequisites of using

• This document is serving as a quick lookup table for the community training/finetuning result, with various language support.
• The models in this repository are open source and are based on voluntary contributions from contributors.
• The use of models must be conditioned on respect for the respective creators. The convenience brought comes from their efforts.

Welcome to share here

• Have a pretrained/finetuned result: model checkpoint (pruned best to facilitate inference, i.e. leave only ema_model_state_dict) and corresponding vocab file (for tokenization).
• Host a public huggingface model repository and upload the model related files.
• Make a pull request adding a model card to the current page, i.e. src\f5_tts\infer\SHARED.md.

Supported Languages

• Multilingual
  • F5-TTS v1 v0 Base @ zh & en @ F5-TTS
• English
• Finnish
  • F5-TTS Base @ fi @ AsmoKoskinen
• French
  • F5-TTS Base @ fr @ RASPIAUDIO
• Hindi
  • F5-TTS Small @ hi @ SPRINGLab
• Italian
  • F5-TTS Base @ it @ alien79
• Japanese
  • F5-TTS Base @ ja @ Jmica
• Mandarin
• Russian
  • F5-TTS Base @ ru @ HotDro4illa
• Spanish
  • F5-TTS Base @ es @ jpgallegoar

Multilingual

F5-TTS v1 v0 Base @ zh & en @ F5-TTS

Model	🤗Hugging Face	Data (Hours)	Model License
F5-TTS v1 Base	ckpt & vocab	Emilia 95K zh&en	cc-by-nc-4.0

Model: hf://SWivid/F5-TTS/F5TTS_v1_Base/model_1250000.safetensors
# A Variant Model: hf://SWivid/F5-TTS/F5TTS_v1_Base_no_zero_init/model_1250000.safetensors
Vocab: hf://SWivid/F5-TTS/F5TTS_v1_Base/vocab.txt
Config: {"dim": 1024, "depth": 22, "heads": 16, "ff_mult": 2, "text_dim": 512, "conv_layers": 4}

Model	🤗Hugging Face	Data (Hours)	Model License
F5-TTS Base	ckpt & vocab	Emilia 95K zh&en	cc-by-nc-4.0

Model: hf://SWivid/F5-TTS/F5TTS_Base/model_1200000.safetensors
Vocab: hf://SWivid/F5-TTS/F5TTS_Base/vocab.txt
Config: {"dim": 1024, "depth": 22, "heads": 16, "ff_mult": 2, "text_dim": 512, "text_mask_padding": False, "conv_layers": 4, "pe_attn_head": 1}

Other infos, e.g. Author info, Github repo, Link to some sampled results, Usage instruction, Tutorial (Blog, Video, etc.) ...

English

Finnish

F5-TTS Base @ fi @ AsmoKoskinen

Model	🤗Hugging Face	Data	Model License
F5-TTS Base	ckpt & vocab	Common Voice, Vox Populi	cc-by-nc-4.0

Model: hf://AsmoKoskinen/F5-TTS_Finnish_Model/model_common_voice_fi_vox_populi_fi_20241206.safetensors
Vocab: hf://AsmoKoskinen/F5-TTS_Finnish_Model/vocab.txt
Config: {"dim": 1024, "depth": 22, "heads": 16, "ff_mult": 2, "text_dim": 512, "text_mask_padding": False, "conv_layers": 4, "pe_attn_head": 1}

French

F5-TTS Base @ fr @ RASPIAUDIO

Model	🤗Hugging Face	Data (Hours)	Model License
F5-TTS Base	ckpt & vocab	LibriVox	cc-by-nc-4.0

Model: hf://RASPIAUDIO/F5-French-MixedSpeakers-reduced/model_last_reduced.pt
Vocab: hf://RASPIAUDIO/F5-French-MixedSpeakers-reduced/vocab.txt
Config: {"dim": 1024, "depth": 22, "heads": 16, "ff_mult": 2, "text_dim": 512, "text_mask_padding": False, "conv_layers": 4, "pe_attn_head": 1}

• Online Inference with Hugging Face Space.
• Tutorial video to train a new language model.
• Discussion about this training can be found here.

Hindi

F5-TTS Small @ hi @ SPRINGLab

Model	🤗Hugging Face	Data (Hours)	Model License
F5-TTS Small	ckpt & vocab	IndicTTS Hi & IndicVoices-R Hi	cc-by-4.0

Model: hf://SPRINGLab/F5-Hindi-24KHz/model_2500000.safetensors
Vocab: hf://SPRINGLab/F5-Hindi-24KHz/vocab.txt
Config: {"dim": 768, "depth": 18, "heads": 12, "ff_mult": 2, "text_dim": 512, "text_mask_padding": False, "conv_layers": 4, "pe_attn_head": 1}

• Authors: SPRING Lab, Indian Institute of Technology, Madras
• Website: https://asr.iitm.ac.in/

Italian

F5-TTS Base @ it @ alien79

Model	🤗Hugging Face	Data	Model License
F5-TTS Base	ckpt & vocab	ylacombe/cml-tts	cc-by-nc-4.0

Model: hf://alien79/F5-TTS-italian/model_159600.safetensors
Vocab: hf://alien79/F5-TTS-italian/vocab.txt
Config: {"dim": 1024, "depth": 22, "heads": 16, "ff_mult": 2, "text_dim": 512, "text_mask_padding": False, "conv_layers": 4, "pe_attn_head": 1}

• Trained by Mithril Man
• Model details on hf project home
• Open to collaborations to further improve the model

Japanese

F5-TTS Base @ ja @ Jmica

Model	🤗Hugging Face	Data (Hours)	Model License
F5-TTS Base	ckpt & vocab	Emilia 1.7k JA & Galgame Dataset 5.4k	cc-by-nc-4.0

Model: hf://Jmica/F5TTS/JA_21999120/model_21999120.pt
Vocab: hf://Jmica/F5TTS/JA_21999120/vocab_japanese.txt
Config: {"dim": 1024, "depth": 22, "heads": 16, "ff_mult": 2, "text_dim": 512, "text_mask_padding": False, "conv_layers": 4, "pe_attn_head": 1}

Mandarin

Russian

F5-TTS Base @ ru @ HotDro4illa

Model	🤗Hugging Face	Data (Hours)	Model License
F5-TTS Base	ckpt & vocab	Common voice	cc-by-nc-4.0

Model: hf://hotstone228/F5-TTS-Russian/model_last.safetensors
Vocab: hf://hotstone228/F5-TTS-Russian/vocab.txt
Config: {"dim": 1024, "depth": 22, "heads": 16, "ff_mult": 2, "text_dim": 512, "text_mask_padding": False, "conv_layers": 4, "pe_attn_head": 1}

• Finetuned by HotDro4illa
• Any improvements are welcome

Spanish

F5-TTS Base @ es @ jpgallegoar

Model	🤗Hugging Face	Data (Hours)	Model License
F5-TTS Base	ckpt & vocab	Voxpopuli & Crowdsourced & TEDx, 218 hours	cc0-1.0

• @jpgallegoar GitHub repo, Jupyter Notebook and Gradio usage for Spanish model.