Robust Speech Recognition via Large-Scale Weak Supervision
Radford et al., 2023, Robust Speech Recognition via Large-Scale Weak Supervision (Proceedings of the 40th International Conference on Machine Learning (ICML 2023), PMLR 202)
Subscriber deck — free during early access
1 cards in “3.2 Evaluation Metrics” · whole deck
3.2 Evaluation Metrics
- Why did Whisper’s authors normalize text before computing WER?WER penalizes every difference from the reference, including harmless formatting differences, which hits zero-shot models hard since they never saw a dataset’s transcription style. Their normalizer removes such non-semantic differences, with the caveat that it might be overfit to Whisper’s style.
- Radford et al., 2023, Robust Speech Recognition via Large-Scale Weak Supervision (Proceedings of the 40th International Conference on Machine Learning (ICML 2023), PMLR 202), §3.2
Source
- Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey and Ilya Sutskever. 2023. Robust Speech Recognition via Large-Scale Weak Supervision. In Proceedings of the 40th International Conference on Machine Learning (ICML 2023), PMLR 202, pp. 28492–28518. https://proceedings.mlr.press/v202/radford23a.html