Sakhanda Wire
NVDA $152.58 -2.28% MSFT $506.11 -2.29% GOOGL $279.81 -1.38% META $619.07 -1.17% AMZN $256.55 -0.47%
← Back to the news

Nvidia releases free model for real-time speaker identification

Nvidia has released Nemotron 3 Diarization, an AI model designed to determine which person is speaking at each moment of a conversation. The model contains about 100 million parameters, and its weights are freely available. It can distinguish as many as eight speakers and recognize overlapping speech, when multiple people talk simultaneously. Performance declines with more participants, strong background noise, or reverberation.

The system supports both recorded and live audio. When combined with a speech-recognition model such as Parakeet, it can generate transcripts with speaker labels, although the labels remain anonymous, such as “speaker_2.” Users can select audio buffers from four settings, ranging from 30.4 seconds to 0.32 seconds; shorter buffers generally reduce accuracy.

In VoiceArena’s Diarization Benchmark v1, Nemotron 3 recorded a diarization error rate of 14.7%, compared with 19.3% for the next-best system. The benchmark counts overlapping speech and small timing errors at speaker changes. Nvidia’s model also outperformed Streaming Sortformer by an average of 41% across eight test scenarios when using a 1.04-second buffer.

Written by Sakhanda Wire AI from reporting by The Decoder. Facts, quotes and full context are in the original article below.

Originally published by The Decoder on

Read the original on The Decoder ↗

The reporting and the image belong to The Decoder and are credited and linked above; the text on this page is Sakhanda Wire's own summary.

← Back to the news

More stories

All the latest news