NVIDIA releases Nemotron 3.5 Transcribe: 40-language ASR with prompt-controlled transcription
aigclink · x · 2026-10-29
NVIDIA released Nemotron 3.5 Transcribe, a 30B MoE speech model (4.3B active):
- Transcribes 40 languages, separates speakers, handles up to 1-hour recordings
- Can translate, detect/redact sensitive info, and understand audio content
- Prompt-controlled processing: e.g. "add speaker labels, translate to Spanish, redact PII"
- Supports custom vocabulary (names, products, domain terms) for better accuracy
Well suited for meeting notes, interviews, podcasts — producing speaker-labeled transcripts from raw audio. Available on Hugging Face.
Related event: NVIDIA Open-Sources Nemotron 3.5 Transcribe Speech Model(17 posts)→
More from Models
- Anthropic: removing 'think carefully' prompts cut Opus 5.5 latency from 8.2s to 2.1s — FinanceYF5 · 2026-10-29
- Anthropic releases Claude Opus 5.5 prompting guide with 11 best practices — FinanceYF5 · 2026-10-29
- Particle-1 inference goes live on OpenRouter, claiming 28-52% fewer output tokens — ycombinator · 2026-10-29
- Qwen3.8-27B EXL3 quant hits lowest KLD of 0.0002 in just 23 GB — suprjami · 2026-10-29
- Codex Pro user: 6.1 Terra keeps stopping mid-task, breaking cron-driven agent jobs — HunnyMal · 2026-10-29
- Leak: Anthropic's Fable 5.5 reportedly lands Nov 3 or Nov 10 — imjustnewatai · 2026-10-29