AssemblyAI Launches Universal-3.5 Real-Time Transcription
ArtificialAnlys · x · 2026-07-06
AssemblyAI has released Universal-3.5 Pro Realtime, an upgrade from Universal-3 Pro Realtime. The streaming speech-to-text model achieves 4.1% WER on AA-WER Streaming, delivering final results for the first sentence in about 0.4 seconds. It allows context injection at the start of a call and after each agent dialogue turn without requiring a reconnection. It offers three modes: Balanced, Max Accuracy, and Min Latency.
Related event: AssemblyAI Launches Universal-3.5 Speech Models(4 posts)→
More from Multimodal
- Full animation short made with Google Gemini tools, 10,564 frames — AI_Andrew · 2026-10-26
- ostris pre-training "Primal VAE": 1-channel latents match modern 8x VAEs via patching — ostrisai · 2026-10-26
- Slopcore: AI-generated meme films turn low-fi slop aesthetics into a subculture — Yamapama · 2026-10-26
- TischEins, a large Midjourney SREF library, relaunches with new design after security outage — tisch_eins · 2026-10-26
- MOSS-Image-SFT 4-step distillation tops 16.5K downloads on Hugging Face — _akhaliq · 2026-10-26
- Claude Code can now edit videos: a 10-prompt workflow from raw footage to final cut — TawohAwa · 2026-10-26