Automatic Speech Recognition
Welsh
whispercpp
DewiBrynJones's picture
Update README.md
50cf17c verified
metadata
license: apache-2.0
datasets:
  - techiaith/commonvoice_18_0_cy
language:
  - cy
base_model:
  - openai/whisper-base
pipeline_tag: automatic-speech-recognition
tags:
  - whispercpp

whisper-base-ft-cv-cy-cpp

This model is a version of the openai/whisper-base model, fine-tuned on the techiaith/commonvoice_18_0_cy dataset, and then converted for use in whisper.cpp. Whispercpp is a C/C++ port of Whisper that provides high performance inference on offline hardware such as desktops, laptops and mobile devices.

The model is a smaller in size to the corresponding cloud hosted model techiaith/whisper-large-v3-ft-cv-cy. It achieves the following WER results for transcribing:

  • Wer: 42.68
  • Cer: 14.14

Usage

whispercpp makes it easy to use models in many platforms and applications. See the 'examples' folder in the whispercpp github repo for more information and example code.

To get quickly started with whispercpp's basic usage however, follow the 'Quick Start' but download this model with the following command:

$ wget https://huggingface.co/techiaith/whisper-base-ft-cv-cy-cpp/resolve/main/ggml-model.bin