RNN Transducer
A speech recognition architecture that uses an encoder for audio and a prediction network for text, combining them to stream transcriptions live without wait...
Think of It Like This
Like a court stenographer listening to the current syllable while remembering the sentence so far to lock in the correct word instantly.
An RNN Transducer is a type of sequence-to-sequence model optimized for streaming Automatic Speech Recognition (ASR). It emits text incrementally as the audio arrives, completely eliminating the need to wait for the end of the sentence.