librispeech

One dataset says "$20," another says "twenty dollars." Fine-tuning a speech model means finding out which convention wins — and why it matters.

WER is the number everyone reports for speech recognition — and the one that agrees with human judgment least. Here's what it actually measures.

NVIDIA Parakeet beats Whisper on phone dictation despite having more parameters. Here's why architecture matters more than model size.

Does higher-quality audio make AI transcription worse? New research reveals a surprising "sweet spot" for bitrate, challenging a core assumption of...