Scaling up online speech recognition using ConvNets – Facebook Research

Scaling up online speech recognition using ConvNets – Facebook Research

Scaling up online speech recognition using ConvNets – Facebook Research

We design an online end-to-end speech recognition system based on Time-Depth Separable (TDS) convolutions and Connectionist Temporal Classification (CTC). 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.

Source: research.fb.com/publications/scaling-up-online-speech-recognition-using-convnets/