Speech / LibriSpeech test-clean
Speech transcription
English speech recording → transcript
Model size vs score
1 compatible measurements. Parameter counts come from the architecture instantiated from each pinned model configuration; no weights are loaded for this count. Click a point to inspect its evidence.
Measured results
Pilot comparison · first 32 valid examples · wer lower is better. No global SOTA claim.
| System | Score | Samples | Elapsed¹ | Evidence |
|---|---|---|---|---|
| openai/whisper-tiny.en 87c7102498dc | 5.1% | 32 | 5s | Inspect run → 16 Oct 2026, 23:28 UTC |
¹ Evaluation elapsed time includes model loading after dataset selection. It is not serving latency or a fair speed comparison.
What this measures
Benchmark origin
Dataset / split
openslr/librispeech_asr / test
Immutable dataset revision
71cacbfb7e2354c4226d01e70d77d5fca3d04ba1Upstream license metadata
cc-by-4.0 · consult upstream terms
Protocol
codesota-pilot-v1 / asr
Selection
First 32 valid examples in the pinned split order. SNLI rows without a valid label are excluded.
Configuration
Seed 42. Greedy generation for math and code. Mean answer-token log likelihood for text classification, completion and declarative multiple-choice recipes. No model judge.
Planned runs
1 qualified visits in 30 days · priority 42.0. Interest changes execution order, within the approved baseline list.
| System | Status |
|---|---|
| openai/whisper-tiny.en | completed |