Turning Whisper into Real-Time Transcription System
This enables real-time transcription services for applications like multilingual conferences, though it is incremental as it builds on an existing model.
The paper tackled the problem of adapting Whisper, a state-of-the-art multilingual speech recognition model, for real-time transcription by developing Whisper-Streaming, which achieved a latency of 3.3 seconds on long-form speech.
Whisper is one of the recent state-of-the-art multilingual speech recognition and translation models, however, it is not designed for real time transcription. In this paper, we build on top of Whisper and create Whisper-Streaming, an implementation of real-time speech transcription and translation of Whisper-like models. Whisper-Streaming uses local agreement policy with self-adaptive latency to enable streaming transcription. We show that Whisper-Streaming achieves high quality and 3.3 seconds latency on unsegmented long-form speech transcription test set, and we demonstrate its robustness and practical usability as a component in live transcription service at a multilingual conference.