Abstract
With the expanding development of on-device artificial intelligence, voice-enabled devices such as smart speakers, wearables, and other on-device or edge processing systems have been proposed. However, building or obtaining large training datasets that are essential for robust keyword spotting (KWS) remains cumbersome. To address this problem, we propose a deep neural network that can rapidly establish a high-performance KWS system from arbitrary keyword instruction sets. We use an encoder pretrained with a large-scale speech corpus as the backbone network and then design an effective transfer network for KWS. To demonstrate the feasibility of the proposed network, various experiments were conducted on Google Speech Command Datasets V1 and V2. In addition, to verify the applicability of the network for different languages, we conducted experiments using three different Korean speech command datasets. The proposed network outperforms state-of-the-art deep neural networks in both experiments. Furthermore, the proposed network can understand real human voice even when trained with synthetic text-to-speech data.
Highlights
Deep learning has enabled the application of automatic speech recognition (ASR) to commercial services [1]–[7]
Several computational resources are required to create an ASR model based on neural networks, and powerful graphics processing units may be unavailable for some applications
4) We show that a backbone pretrained on an English speech corpus can be used for keyword spotting (KWS) in other languages such as Korean by applying transfer learning
Summary
Deep learning has enabled the application of automatic speech recognition (ASR) to commercial services [1]–[7]. 3) We show that high-performance KWS can be achieved with a small speech command dataset by using text-tospeech (TTS) synthesized data. 4) We show that a backbone pretrained on an English speech corpus can be used for KWS in other languages such as Korean by applying transfer learning. A multi-head attention recurrent neural network has achieved a high accuracy of 97.2% and 98.0% on Google Speech Command Datasets V1 and V2 [21], respectively, overcoming the gap in model performance between streaming KWS and test datasets. The weight of the diversity loss is adjusted by its hyperparameters
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.