Recently, the speech community is seeing a significant trend of moving from deep neural
network based hybrid modeling to end-to-end (E2E) modeling for automatic speech …

End-to-end automatic speech recognition (ASR) has become an increasingly popular area
of research, with two main models being online and offline ASR. Online models aim to …

In this paper, we propose a dynamic cascaded encoder Automatic Speech Recognition
(ASR) model, which unifies models for different deployment scenarios. Moreover, the model …

Transformer-based end-to-end speech recognition has achieved great success. However,
the large footprint and computational overhead make it difficult to deploy these models in …

Streaming ASR with strict latency constraints is required in many speech recognition
applications. In order to achieve the required latency, streaming ASR models sacrifice …

We present a streaming, Transformer-based end-to-end automatic speech recognition
(ASR) architecture which achieves efficient neural inference through compute cost …

Neural contextual biasing for end-to-end neural ASR transducers has shown significant
improvements in the recognition of named entities, such as contact names or device names …

Although end-to-end (E2E) automatic speech recognition (ASR) has shown state-of-the-art
recognition accuracy, it tends to be implicitly biased towards the training data distribution …

We introduce Caching Networks (CachingNets), a speech recognition network architecture
capable of delivering faster, more accurate decoding by leveraging common speech …

Varying-size models are often required to deploy ASR systems under different hardware
and/or application constraints such as memory and latency. To avoid redundant training and …