TheMachine Press

A daily newspaper for the age of artificial intelligence.

Morning editionPermanent story

research

Speech Interfaces Got a Common Information Scale

Open-vocabulary mutual information compares neural speech decoders even when their datasets, vocabularies and recording setups differ.

Published Updated Story ID: mp-2026-09-03-027
Read the complete editionStory JSON

Summary

Open-vocabulary mutual information compares neural speech decoders even when their datasets, vocabularies and recording setups differ.

Accuracy and word error rate can overstate communication when they score only the words a brain-computer interface already supports. OVMI instead measures conveyed information against a reference distribution of what a user may wish to say, putting coverage and decoding accuracy on the same scale. Applied to existing systems, it exposed vocabulary tradeoffs and found that selecting a vocabulary to maximize OVMI improved accuracy by as much as 16.3% across three speech domains.

Why it matters

Open-vocabulary mutual information compares neural speech decoders even when their datasets, vocabularies and recording setups differ.

Limits and context

  • Accuracy and word error rate can overstate communication when they score only the words a brain-computer interface already supports.

Key claims

  1. Open-vocabulary mutual information compares neural speech decoders even when their datasets, vocabularies and recording setups differ.

    Qualification: Accuracy and word error rate can overstate communication when they score only the words a brain-computer interface already supports.

    Evidence: source-2026-09-03-016

Sources

  1. arXiv preprint 2609.02887arXiv · primary research

Corrections

No corrections have been recorded for this story.