Since 2020, Estonia's Riigikogu has used HANS, an AI speech-recognition tool built with Tallinn University of Technology, to auto-transcribe plenary debates — targeting 93–95% accuracy and publishing machine-readable verbatim records within an hour of each sitting.
93–95%
Target transcription accuracy for plenary sittings
~1,500 hours
Training data volume
2 years
Development period
Details
Maturity
Established
Promoter
Riigikogu (Parliament of Estonia), with Tallinn University of Technology (TalTech) and Finestmedia
Period
2020–present
Keywords
speech recognition, parliamentary transcription, open data, legislative support
Context
HANS is an AI speech-recognition pipeline introduced by Estonia's Riigikogu in autumn 2020, built by Finestmedia using speech-recognition technology from Tallinn University of Technology and trained on roughly 1,500 hours of recordings and transcripts, to auto-transcribe plenary debates into open, machine-readable verbatim records — with human editors reviewing each draft — aiming to publish within an hour of a sitting ending.
Implementation
Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.
Data sources
Where this practice's information was retrieved from, and when.
Chile's Chamber of Deputies built CAMINAR, a four-part generative-AI suite for bill drafting, transcription and regulatory-impact review, using closed authorised …
