Very Fast Keyword Spotting System with Real Time Factor below 0.01
This work addresses the need for efficient keyword spotting in speech processing, though it appears incremental as it builds on existing neural network methods with specific optimizations.
The authors tackled the problem of building a very fast keyword spotting system by proposing optimizations across all steps of the design, achieving a real-time factor as low as 0.001 with GPU acceleration.
In the paper we present an architecture of a keyword spotting (KWS) system that is based on modern neural networks, yields good performance on various types of speech data and can run very fast. We focus mainly on the last aspect and propose optimizations for all the steps required in a KWS design: signal processing and likelihood computation, Viterbi decoding, spot candidate detection and confidence calculation. We present time and memory efficient modelling by bidirectional feedforward sequential memory networks (an alternative to recurrent nets) either by standard triphones or so called quasi-monophones, and an entirely forward decoding of speech frames (with minimal need for look back). Several variants of the proposed scheme are evaluated on 3 large Czech datasets (broadcast, internet and telephone, 17 hours in total) and their performance is compared by Detection Error Tradeoff (DET) diagrams and real-time (RT) factors. We demonstrate that the complete system can run in a single pass with a RT factor close to 0.001 if all optimizations (including a GPU for likelihood computation) are applied.