Joy Dataset: 100,000 Hours of Dual-Channel Conversations for Voice Models
Aiden_Tech_Ai · x · 2026-10-30
shobhitbanga introduced Joy, a dual-channel conversational dataset: real, unscripted conversations between two people, each on a separate audio track. Scale: 100,000 hours, 50,000 speakers, 10 languages. It's built to teach voice models how people actually talk — when to listen, when to jump in, when to say "hmm". The retweeter notes that on a 30-second call they might no longer catch a well-trained agent.
Related event: Joy Dataset Released: 100K Hours of Dual-Channel Human Dialogue(2 posts)→
More from Research
- One-chart breakdown: Mistral Large 4 vs DeepSeek, GLM and six other large MoE architectures — eliebakouch · 2026-10-30
- AI-driven scientific discoveries cost $10M+ compute on Navier-Stokes, $100M+ overall — kimmonismus · 2026-10-30
- E2E-SWE benchmark: 186 tasks test coding agents building repos from scratch, best pass@1 just 67.7% — jyangballin · 2026-10-30
- LeRoPE makes RoPE frequencies learnable, beating RoPE across 52M-2.5B scales — eliebakouch · 2026-10-30
- ML Commons launches 'benchmarks as a service' to kill benchmaxxing — RexDouglass · 2026-10-30
- Redwood's Screwtape: 35 monitor-evading attacks from 252 runs on real agent transcripts — dl_weekly · 2026-10-30