See who's speaking and when, live from your microphone
Made by NVIDIA
View profile on Hugging Face (opens in a new tab)Live Speaker Diarization is Nvidia's browser demo for figuring out “who spoke when” in audio: start your microphone or upload a short recording, and it labels overlapping speaker turns in near-real time, showing speaker-activity lanes alongside a live transcript.
A separate multilingual mode combines Nvidia's streaming speech-recognition model with its diarization model and auto-detects the spoken language. The public demo caps live-mic sessions at 30 seconds and uploaded files at 2 minutes, keeping it usable as a free shared demo rather than a full meeting-transcription tool.
You could use it to…
- Start your microphone and see who's speaking labeled live
- Get a multilingual mode that auto-detects the spoken language
- Try Nvidia's own demo with a 30-second live-mic time limit
Opens on Hugging Face. BuildTube has not run or verified it.
Can I use this?
You'll need Just a web browser and a microphone · nothing to install
Worth knowing The public demo enforces short time limits (30 seconds live, 2 minutes per file) that would not apply to Nvidia's underlying model outside this shared demo, so it's a capability preview rather than a full transcription tool. BuildTube has not run or verified this software.
Why it's here
Added 25 September
New Sept 1 · 53 likes on Hugging Face by Sept 25
- 53 people have liked it on Hugging Face.
Numbers checked on 25 September 2026; not refreshed since.
Behind it
See the code on Hugging FaceBuildTube has not run or verified this project. Everything above is written from what the creator published.
Made this?