Preprint
Better Transcription of UK Supreme Court Hearings
Hadeel Saadany et al., arXiv (Cornell University), 2022
Read the originalRead on October 16, 2026
- What it is
- A research and industry project on transcribing UK Supreme Court hearings. The authors adapt a general speech recognition model with a court-specific language model and common legal phrases, and report that this improves word error rate and avoids critical errors on legal jargon.
- How much weight it carries
- Preprint, not peer-reviewed in the version we read. It studies British court audio, so results may not carry to US law firms or to one lawyer dictating. The abstract is the part we could read and quote, and it reports no single headline error rate.
Facts taken from this source
- The 2022 study reports that adding an in-domain language model and common legal phrases to a generic speech model can improve its word error rate on court hearings.
“fine-tuning a generic off-the-shelf pre-trained Automatic Speech Recognition (ASR) system with an in-domain language model as well as infusing common phrases extracted with a collocation detection model can improve not only the Word Error Rate (WER)” (Abstract)
- The authors of a 2022 study on UK Supreme Court hearings report that adapting a generic speech model can avoid critical errors specific to legal jargon and terminology used in British courts.
“avoid critical errors that are specific of the legal jargon and terminology commonly used in British courts” (Abstract)