Mickael Rouvier

WER hides the real story: new metrics reveal how language model rescoring in ASR impacts grammatical correctness and semantic accuracy.

Current ASR metrics, even those leveraging embeddings, fail to align with human perception of transcription quality, as revealed by a new human-annotated dataset.

LLMs can judge speech recognition quality with near-human accuracy, blowing away traditional metrics like Word Error Rate.

Multi-corpus training can actually *hurt* spoofing detection, unless you strip out dataset-specific biases with this clever domain-invariant feature extraction trick.