Small-Footprint Open-Vocabulary Keyword Spotting with Quantized LSTM Networks
arXiv:2002.10851
Abstract
We explore a keyword-based spoken language understanding system, in which the intent of the user can directly be derived from the detection of a sequence of keywords in the query. In this paper, we focus on an open-vocabulary keyword spotting method, allowing the user to define their own keywords without having to retrain the whole model. We describe the different design choices leading to a fast and small-footprint system, able to run on tiny devices, for any arbitrary set of user-defined keywords, without training data specific to those keywords. The model, based on a quantized long short-term memory (LSTM) neural network, trained with connectionist temporal classification (CTC), weighs less than 500KB. Our approach takes advantage of some properties of the predictions of CTC-trained networks to calibrate the confidence scores and implement a fast detection algorithm. The proposed system outperforms a standard keyword-filler model approach.
References in corpus (11)
- Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition
- Pyroomacoustics: A Python package for audio room simulations and array processing algorithms
- Achieving Human Parity in Conversational Speech Recognition
- Hello Edge: Keyword Spotting on Microcontrollers
- End-to-End ASR-free Keyword Search from Speech
- Effective Quantization Methods for Recurrent Neural Networks
- Online Keyword Spotting with a Character-Level Recurrent Neural Network
- An End-to-End Architecture for Keyword Spotting and Voice Activity Detection
- Sequence Discriminative Training for Deep Learning based Acoustic Keyword Spotting
- DONUT: CTC-based Query-by-Example Keyword Spotting
- Automatic Grammar Augmentation for Robust Voice Command Recognition