Hume Audio Decision API
Hume AI · Released 2 Nov 2026
An audio-native decision API that answers developer-defined questions about speech and sound. Raw audio goes in; structured answers with confidence scores come out. Question types include yes/no, multiple choice, rating scales, and time spans. Aimed at live calls, contact-center scoring, and search over audio libraries.
- Maker
- Hume AI
- Release date
- 2 Nov 2026
- First seen
- 2 Nov 2026
- Access
- Playground for realtime and file-upload trials; API access by waitlist, as stated in the announcement.
- Pricing (as publicly stated)
- A public list price was not stated in the 2 Nov 2026 announcement.
- Modalities
- audio
- Question types
- yes/no, choice, scale, span
- Latency (as stated)
- Hume said it is fast enough to ask dozens of questions every few seconds on a live call, and to search an archive of 40,000 recordings in minutes.
- Official announcement
- https://www.hume.ai/blog/introducing-hume-audio-decision-api
Benchmark claims
These are claims as published by the maker or in the linked official post. We have not independently rerun every benchmark.
Hume said that in an initial evaluation, its Audio Decision model achieved 70.7% agreement with a human panel's average judgment, compared with 56.5% for the individual-rater baseline.
Names and marks belong to their owners. This page is an independent summary, not a product page from Hume AI.