Instructions to use argmaxinc/whisperkit-coreml with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- WhisperKit
How to use argmaxinc/whisperkit-coreml with WhisperKit:
# Install CLI with Homebrew on macOS device brew install whisperkit-cli # View all available inference options whisperkit-cli transcribe --help # Download and run inference using whisper base model whisperkit-cli transcribe --audio-path /path/to/audio.mp3 # Or use your preferred model variant whisperkit-cli transcribe --model "large-v3" --model-prefix "distil" --audio-path /path/to/audio.mp3 --verbose
- Notebooks
- Google Colab
- Kaggle
Upload model openai_whisper-large-v2_turbo_955MB - QLoRA compressed encoder decoder variant
046ab9f verified Download openai_whisper-large-v2_turbo_955MB/TextDecoder.mlmodelc/coremldata.bin from argmaxinc/whisperkit-coreml: direct link, hf CLI and curl.
- Browser
- Download file 637 Bytes
-
https://huggingface.co/argmaxinc/whisperkit-coreml/resolve/5e296a6710b640f6c0cb739c5b6f8c8a489ed5cc/openai_whisper-large-v2_turbo_955MB/TextDecoder.mlmodelc/coremldata.bin
- Command line
-
hf download hf://argmaxinc/whisperkit-coreml@5e296a6710b640f6c0cb739c5b6f8c8a489ed5cc/openai_whisper-large-v2_turbo_955MB/TextDecoder.mlmodelc/coremldata.bin
-
curl -L -o coremldata.bin https://huggingface.co/argmaxinc/whisperkit-coreml/resolve/5e296a6710b640f6c0cb739c5b6f8c8a489ed5cc/openai_whisper-large-v2_turbo_955MB/TextDecoder.mlmodelc/coremldata.bin
637 Bytes
- Xet hash:
- 483d055dd18b42efe568846593e2db2aff4451febd79e57b0d390ae10f94081b
- Size of remote file:
- 637 Bytes
- SHA256:
- 5050593c9e3117c5e143a6bc3aa256287ae8f3beada904e2339cae7dc457600b
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.