An All-Neural On-Device Speech Recognizer

“Our new all-neural, on-device Gboard speech recognizer is initially being launched to all Pixel phones in American English only.” (Final paragraphs on compression and launch)

“the new recognizer is always available, even when you are offline ... After compression, the final model is 80MB. Our new all-neural, on-device Gboard speech recognizer is initially being launched to all Pixel phones in American English only.” (Introduction and Offline Recognition section)

“we present a model trained using RNN transducer (RNN-T) technology that is compact enough to reside on a phone” (Introduction of the post)

“Today, we're happy to announce the rollout of an end-to-end, all-neural, on-device speech recognizer to power speech input in Gboard.” (Opening section of the post)

“The RNN-T we trained offers the same accuracy as the traditional server-based models but is only 450MB” (Offline Recognition section)

“the new recognizer is always available, even when you are offline.” (Post body)

“continuously processes input samples and streams output symbols, a property that is welcome for speech dictation. In our implementation, the output symbols are the characters of the alphabet.” (Body of the 2019 Google Research blog post)

“the search graph remains quite large, almost 2GB for our production models. Since this is not something that could be hosted easily on a mobile phone, this method requires online connectivity to work properly.” (Offline Recognition section)