Skip to content
ModelListedExternal
Download

IndicF5 Kannada Bedtime

Fine-tuned ai4bharat/IndicF5 on Kannada bedtime-story speech.

KindOtherBase modelai4bharat/IndicF5VersionRevision 538cb93

Model source

Kind
Other
File size
1.26 GB
Base model
ai4bharat/IndicF5
Version
vRevision 538cb93
Updated
Sep 13, 2026
Downloads
0
Source
Hugging Face

Source description

IndicF5 Kannada Bedtime Story Fine-Tune

Fine-tuned ai4bharat/IndicF5 on Kannada bedtime-story speech.

Training Details

  • Base model: ai4bharat/IndicF5 (0.4B params)
  • Training data: Rasa Kannada (18 clips, ~2min)
  • Training steps: 300
  • Learning rate: 1e-5
  • Final loss: 0.4553
  • GPU: A100-80GB (Modal)

How to Use

from transformers import AutoModel
import soundfile as sf

model = AutoModel.from_pretrained(
    "mitvho09/IndicF5-Kannada-Bedtime",
    trust_remote_code=True,
)

audio = model(
    "ಮಕ್ಕಳೇ, ನಿದ್ರೆ ಮಾಡಿ.",
    ref_audio_path="reference.wav",
    ref_text="transcript of reference",
)

sf.write("output.wav", audio, 24000)

Evaluation

Compare stock vs fine-tuned audio in the comparison/ directory.

Fine-Tuning Details

This is an experimental fine-tune for the DreamVoice competition badge. Stock IndicF5 may perform better for general Kannada TTS.

Sources

1 source
Hugging FacePrimaryVersion 538cb937e32633bc8ca74857c6f76206370df5a7
huggingface.co/mitvho09/IndicF5-Kannada-Bedtime/tree/538cb937e32633bc8ca74857c6f76206370df5a7

Verified Sep 13

Model artifacts

1 artifact

cfm.pt

pt · 1.26 GB · SHA-256 8efd19b85e33…b761 · Hugging Face

Download file

Source excerpts

2 excerpts

--- language: kn tags: - text-to-speech - kannada - indicf5 - bedtime-stories base_model: ai4bharat/IndicF5 library_name: transformers license: mit --- # IndicF5 Kannada Bedtime Story Fine-Tune Fine-tuned [ai4bharat/IndicF5](https://huggingface.co/ai4bharat/IndicF5) on Kannada bedtime-story speech. ## Training Details - **Base model**: ai4bharat/IndicF5 (0.4B params) - **Training data**: Rasa Kannada (18 clips, ~2min) - **Training steps**: 300 - **Learning rate**: 1e-5 - **Final loss**: 0.4553 - **GPU**: A100-80GB (Modal) ## How to Use ```python from transformers import AutoModel import soundfile as sf model = AutoModel.from_pretrained( "mitvho09/IndicF5-Kannada-Bedtime", trust_remote_code=True, ) audio = model( "ಮಕ್ಕಳೇ, ನಿದ್ರೆ ಮಾಡಿ.", ref_audio_path="reference.wav", ref_text="transcript of reference", ) sf.write("output.wav", audio, 24000) ``` ## Evaluation Compare stock vs fine-tuned audio in the `comparison/` directory. ## Fine-Tuning Details This is an experimental fine-tune for the DreamVoice competition badge. Stock IndicF5 may perform better for general Kannada TTS.

Source context: 0 downloads · 1 likes · Pipeline text-to-speech · Library transformers · Repo mitvho09/IndicF5-Kannada-Bedtime