Abstract
Nowadays automatic speech recognition (ASR) systems can achieve higher and higher accuracy rates depending on the methodology applied and datasets used. The rate decreases significantly when the ASR system is being used with a non-native speaker of the language to be recognized. The main reason for this is specific pronunciation and accent features related to the mother tongue of that speaker, which influence the pronunciation. At the same time, an extremely limited volume of labeled non-native speech datasets makes it difficult to train, from the ground up, sufficiently accurate ASR systems for non-native speakers.In this research, we address the problem and its influence on the accuracy of ASR systems, using the style transfer methodology. We designed a pipeline for modifying the speech of a non-native speaker so that it more closely resembles the native speech. This paper covers experiments for accent modification using different setups and different approaches, including neural style transfer and autoencoder. The experiments were conducted on English language pronounced by Japanese speakers (UME-ERJ dataset). The results show that there is a significant relative improvement in terms of the speech recognition accuracy. Our methodology reduces the necessity of training new algorithms for non-native speech (thus overcoming the obstacle related to the data scarcity) and can be used as a wrapper for any existing ASR system. The modification can be performed in real time, before a sample is passed into the speech recognition system itself.
Highlights
Automatic speech recognition is a function that has been the subject of extensive research for decades
The LibriSpeech test-clean dataset was used for evaluating the Time Delay Neural Network (TDNN)-based automatic speech recognition (ASR) network we trained, and the result achieved in our test was 10% Character Error Rate (CER) and 12.5% Word Error Rate (WER)
5 Conclusions In this research, we explained the problem of non-native speech recognition and the reason why training ASR systems adapted for such speech may be problematic
Summary
Automatic speech recognition is a function that has been the subject of extensive research for decades. Developed speech recognition tools can recognize speech with an almost human-like accuracy, depending on the dataset and benchmark test used [1]. Such performance can be achieved only when the system is used for recognizing the speech of native speakers (i.e., the native speakers of the language represented by the dataset used to train the ASR system). The main reason for this drop is the presence of patterns related to the speaker’s mother tongue which can influence the pronunciation of the second language. This makes their language biased to some extent which causes
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: EURASIP Journal on Audio, Speech, and Music Processing
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.