We Help Voice AI Handle Group Conversations
Details
- External ID
- 48649105
- Source
- HN
- Company
- —
- Product
- We Help Voice AI Handle Group Conversations
- Website domain
- github.com
- Launched
- June 23, 2026
- Cohort
- —
- Upvotes
- 5
- Upvotes percentile
- 0.12568306010928962
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Hey folks. We built SAA (Selective Auditory Attention) after trying to find ways to make a good experience with multiple robots/multiple agents. What typically ended up happening is they'd never stop talking.This is an SDK you can put before your STT. It lets you know when your device is being spoken to or not without a wakeword. You can use it for: -Single AI, Multi human -Multi AI, Single human -Multi AI, Multi human (we recommend also adding a wakeword on top for a better system)There are two models. One that is video + audio and one that is just audio. The way it overall works is that it looks for shifts in attention patterns (body language changes, vocal patterns) to work. It's a tough problem to nail as every human being is different in how they interact with people/devices.Let me know how it is!
Enrichment
- Theme
- ai agents for calls and meetings
- Vertical
- Horizontal
- Function
- Agent / copilot
- Audience
- B2B
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- voice ai for group conversations
- Manually corrected
- False
Could you build this?
No Selective auditory attention for multi-party voice AI requires low-latency digital signal processing (DSP), beamforming, speaker diarization, or specialized neural audio filtering running prior to STT.
What it would actually take: A real version requires an audio processing pipeline running locally on embedded devices or edge nodes using C++/Rust and WebRTC/low-level audio buffers. It leverages trained neural audio separation models (e.g., Conv-TasNet or target sound extraction) combined with multi-microphone array direction-of-arrival (DoA) algorithms to filter speech directed specifically at the agent under 50ms latency. This requires deep acoustic engineering, digital signal processing, and low-latency systems programming.
Discussion
No comments on this launch.
Competitors
Other products that read as similar to this one — 248 launches clear the similarity bar, closest 8 shown.
Attention rank: #238 of 249 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 236 days after the earliest competitor.
- Talkasu: Rehearse Hard Talks · ph · 2026-09-23 · 2 upvotes · similarity 0.47
- I built a voice AI that responds like a real woman · hn · 2026-03-25 · 6 upvotes · similarity 0.47
- Multimodal perception system for real-time conversation · hn · 2026-02-10 · 54 upvotes · similarity 0.47
- AI Group Call · ph · 2026-08-10 · 193 upvotes · similarity 0.46
- Sparrow-2 · hn · 2026-09-08 · 11 upvotes · similarity 0.46
- I built an AI conversation partner to practice speaking languages · hn · 2026-01-30 · 65 upvotes · similarity 0.45
- Hola AI · ph · 2026-09-22 · 94 upvotes · similarity 0.44
- Hush · ph · 2026-06-23 · 191 upvotes · similarity 0.43
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a agent / copilot tool for Agriculture yet.