Amazon Bedrock Data Automation Adds Custom Vocabulary for Domain-Specific Transcriptions

Custom vocabulary support in Amazon Bedrock Data Automation enhances transcription accuracy for specialized terms across diverse industries and multiple languages.

By Isabel Turner · Editorial pseudonym · 2026-04-03 · 1 min read

AI-ASSISTED REPORTING · RETROSPECTIVE ARCHIVE

Retrospective edition: 2026-04-03. Source published: 2026-04-03. Added to this archive: 2026-11-03. This briefing was prepared later from the linked announcement; it was not published on this website on the historical edition date.

Amazon added a Data Automation Library to Bedrock Data Automation, enabling users to supply custom lists of domain-specific words such as medical terms or acronyms to improve speech recognition accuracy in unstructured audio and video content. This capability assists sectors like healthcare, legal, finance, and media in recognizing specialized terminology accurately within GenAI-powered applications.

Supporting 11 languages including English, Spanish, and Chinese variants, the feature also allows specification of preferred display names for recognized terms. Offered at no extra cost, this update is available in multiple AWS regions, aiming to refine transcription quality in sector-specific contexts as documented by AWS.

Source & further reading

Amazon Bedrock Data Automation now supports custom vocabulary ↗

Source published 2026-04-03. This report was prepared with AI assistance from the linked primary-source material. Claims and limitations should be checked against the original announcement or research.