Nemotron Speech Streaming is an NVIDIA-hosted demo that transcribes speech from a microphone or uploaded audio files in real time. It delivers clear, readable text output without delay. The tool showcases NVIDIA's speech recognition models and is available as a Hugging Face Space.
In the Speech to text space, Nemotron Speech Streaming takes a focused approach. It focuses on converting spoken language from live microphone or audio files into accurate text instantly. Nemotron Speech Streaming is an open-source project aimed at developers. It is available for free. It ships for the web, and it can be self-hosted.
NVIDIA builds and maintains Nemotron Speech Streaming. Among its 3 catalogued features are Live Transcription, File Upload, and Real-time Output.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do