Ir para o conteúdo
ModelVerificadoExterno
Baixar

Whisper Base Myst Gclora

Whisper-base adapted to the MyST dataset with GC-LoRA (Gated Convolutional LoRA), from the paper "GC-LoRA: Gated Convolutional LoRA for Parameter-Efficient Acoustic Adaptation" (Interspeech 2026).

Fontehuggingface.co/balaji1312/whisper-base-myst-gclora/tree/be7897f5fc24cae61d9137d8448daec6f88e0106
TipoOtherVersãoRevision be7897fPublicador@balaji1312Cnota

Fonte do modelo

Tipo
Other
Tamanho do arquivo
277 MB
Versão
vRevision be7897f
Atualizado
17 de out. de 2026
Publicador
@balaji1312
Downloads
4
Fonte
Hugging Face

Descrição da fonte

Whisper-base MyST (GC-LoRA)

Whisper-base adapted to the MyST dataset with GC-LoRA (Gated Convolutional LoRA), from the paper "GC-LoRA: Gated Convolutional LoRA for Parameter-Efficient Acoustic Adaptation" (Interspeech 2026).

  • Base model: openai/whisper-base.en
  • Method: GC-LoRA adapter on the encoder attention output projections (rank 8, kernel 31, scaling 16)
  • MyST test WER: 12.2%
  • Code: https://github.com/balaji1312/gc_lora

Usage

The checkpoint bundles the frozen Whisper backbone together with the trained GC-LoRA adapter. Loading requires the custom modeling code in the gc_lora repository; see src/bin/decode_asr.py there for loading and decoding.

Citation

@inproceedings{shankar2026gclora,
  author    = {Shankar, Natarajan Balaji and Wang, Zilai and Zhang, Kaiyuan and Shi, Mohan and Alwan, Abeer},
  title     = {{GC-LoRA}: Gated Convolutional {LoRA} for Parameter-Efficient Acoustic Adaptation},
  booktitle = {Interspeech 2026},
  year      = {2026},
}

Fontes

1 fonte
Hugging FacePrimárioVersão be7897f5fc24cae61d9137d8448daec6f88e0106
huggingface.co/balaji1312/whisper-base-myst-gclora/tree/be7897f5fc24cae61d9137d8448daec6f88e0106

Verificado 26 de set.

Artefatos de modelo

1 artefato

model.safetensors

safetensors · 277 MB · SHA-256 1b5e2c8ab3c0…e8bb · Hugging Face

Baixar

Trechos de fonte

3 trechos
Readme11 de jul.huggingface.co/balaji1312/whisper-base-myst-gclora

--- license: apache-2.0 language: en library_name: transformers pipeline_tag: automatic-speech-recognition tags: - whisper - automatic-speech-recognition - peft - lora - gc-lora - myst --- # Whisper-base MyST (GC-LoRA) Whisper-base adapted to the **MyST** dataset with **GC-LoRA** (Gated Convolutional LoRA), from the paper *"GC-LoRA: Gated Convolutional LoRA for Parameter-Efficient Acoustic Adaptation"* (Interspeech 2026). - **Base model:** `openai/whisper-base.en` - **Method:** GC-LoRA adapter on the encoder attention output projections (rank 8, kernel 31, scaling 16) - **MyST test WER:** **12.2%** - **Code:** https://github.com/balaji1312/gc_lora ## Usage The checkpoint bundles the frozen Whisper backbone together with the trained GC-LoRA adapter. Loading requires the custom modeling code in the [gc_lora repository](https://github.com/balaji1312/gc_lora); see `src/bin/decode_asr.py` there for loading and decoding. ## Citation ```bibtex @inproceedings{shankar2026gclora, author = {Shankar, Natarajan Balaji and Wang, Zilai and Zhang, Kaiyuan and Shi, Mohan and Alwan, Abeer}, title = {{GC-LoRA}: Gated Convolutional {LoRA} for Parameter-Efficient Acoustic Adaptation}, booktitle = {Interspeech 2026}, year = {2026}, } ```

Note26 de set.huggingface.co/balaji1312/whisper-base-myst-gclora

Source context: 4 downloads · 0 likes · Pipeline automatic-speech-recognition · Library transformers · Repo balaji1312/whisper-base-myst-gclora

Note17 de jul.huggingface.co/balaji1312/whisper-base-myst-gclora

Source context: 1 downloads · 0 likes · Pipeline automatic-speech-recognition · Library transformers · Repo balaji1312/whisper-base-myst-gclora