Capability matrix
- Compute and format FP32 · Generated x86 encoder/decoder runtime.
- Local entry point Audio transcription.
- Server Speech HTTP service. Whisper has a dedicated transcription service.
- Harness Audio client. Verify the transcript with a known recording for each size.
Choose the download
Publisher model card · CKE input files
Publisher task: automatic speech recognition. CKE example: FP32. Use this artifact identifier:
hf://openai/whisper-base. Tiny · Small
Get started
Audio uses the audio subcommand; see the v8 runbook for generated artifact directories and language/task options
version/v8/scripts/cks-v8-run audio hf://openai/whisper-base --wav recording.wav --language en --task transcribeWhat has actually run?
Token-exact Hugging Face trajectory on public JFK fixtures. JFK fixture; 33-second Base long-form fixture.
CPU evidence. Generated x86 encoder/decoder runtime.
Performance. For speech, compare transcription time with audio duration and check the transcript against known words. ARM throughput for this exact artifact is unverified here. Benchmark evidence · 32K to 128K results
Can I fine-tune it?
Full-checkpoint training, LoRA and other PEFT methods: no verified CKE recipe for this exact checkpoint in the reviewed evidence. For a trainable starter, use the FP32 and adapter recipes.