README.md

metadata

library_name: aneforge
license: apache-2.0
base_model: openai/whisper-base.en
pipeline_tag: automatic-speech-recognition
tags:
  - aneforge
  - apple-neural-engine

Whisper base.en (ANEForge)

An unmodified duplicate of openai/whisper-base.en, tagged for use with ANEForge so the weights load and run directly on the Apple Neural Engine (no CoreML). Weights are byte-identical to the source; see the original repo for the model details. Licensed apache-2.0 from the source.

Use with ANEForge

import aneforge as af

asr = af.load_whisper("aneforge/whisper-base.en")
print(asr.transcribe(audio))        # 16 kHz mono waveform -> transcript

ANEForge compiles the model's graph into a single ANE program and streams the weights from this repo via huggingface_hub. See the docs and the paper.