Automatic Speech Recognition
Transformers
Safetensors
English
whisper
text-generation-inference
unsloth
Instructions to use Kibalama/lugandaSTT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Kibalama/lugandaSTT with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Kibalama/lugandaSTT")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("Kibalama/lugandaSTT") model = AutoModelForSpeechSeq2Seq.from_pretrained("Kibalama/lugandaSTT") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Unsloth Studio new
How to use Kibalama/lugandaSTT with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Kibalama/lugandaSTT to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Kibalama/lugandaSTT to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for Kibalama/lugandaSTT to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="Kibalama/lugandaSTT", max_seq_length=2048, )
How do I convert the file for whispper.cpp compatibility?
#1
by JLouisBiz - opened
$ python convert-pt-to-ggml.py ~/Downloads/model.safetensors /mnt/nvme0n1/LLM/git/whisper.cpp/ .
/home/data1/protected/venv/lib/python3.13/site-packages/torch/_weights_only_unpickler.py:549: UserWarning: Detected pickle protocol 216 in the checkpoint, which was not the default pickle protocol used by `torch.load` (2). The weights_only Unpickler might not support all instructions implemented by this protocol, please file an issue for adding support if you encounter this.
warnings.warn(
Error: failed to load PyTorch model file: /home/data1/protected/Downloads/model.safetensors
I have tried that way. Any other way to make it whisper.cpp compatible?
I tried it, but it is not compatible with lg language, and it gives wrong Luganda out.
Sorry Louis, It was trained on a little luganda corpus. I'm currently still collecting more data for better training