Wed-3-9-7 Scaling Up Online Speech Recognition Using ConvNets

Vineel Pratap(Facebook), Qiantong Xu(Facebook AI Research), Jacob Kahn(Facebook AI), Gilad Avidov(Facebook AI), Tatiana Likhomanenko(Facebook AI Research), Awni Hannun(Facebook AI Research), Vitaliy Liptchinsky(Facebook AI), Gabriel Synnaeve(Facebook AI Research) and Ronan Collobert(Facebook AI Research)

Abstract: We design an online end-to-end speech recognition system based on Time-Depth Separable (TDS) convolutions and Connectionist Temporal Classification (CTC). We improve the core TDS architecture in order to limit the future context and hence reduce latency while maintaining accuracy. The system has almost three times the throughput of a well tuned hybrid ASR baseline while also having lower latency and a better word error rate. Also important to the efficiency of the recognizer is our highly optimized beam search decoder. To show the impact of our design choices, we analyze throughput, latency, accuracy, and discuss how these metrics can be tuned based on the user requirements.

Paper

prev Wed-3-9-6 Streaming on-device end-to-end ASR system for privacy-sensitive voicetyping

next Wed-3-9-8 Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition

About

About the Conference

Welcome from the Chair

Conference Committees

Calls