AI engineering glossary · Generative media multimodal
What is Streaming ASR?
Streaming ASR: Speech recognition that outputs text while the person is still speaking, using limited look-ahead so latency stays low. It trades some accuracy for responsiveness compared with processing a complete recording.
Streaming ASR: an example
Live subtitles on a cricket commentary stream appear within about half a second of the speaker's words.
Related terms
- RNN-Transducer
- Automatic speech recognition
- Voice activity detection
More Generative media multimodal terms
- StyleGAN
- Super-resolution
- Tacotron
- Temporal consistency
- Text-to-3D generation
- Text-to-image generation
- Text-to-speech
- Text-to-video generation
Where you learn it
Streaming ASR is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.