Search

Cross-lingual speech recognition leverages data from a variety of source languages to train systems capable of identifying speech in a target language, thus promoting intercultural communication and understanding. To address the issues of how to utilize multilingual data to improve the recognition performance of low resource languages in cross-lingual speech recognition, domain shift or interference between source and target languages, task weights and data distribution between different languages, a cross lingual speech recognition model was studied through feature prompts. To simplify the traditional process of requiring professionals to label phonemes uniformly, a cross-lingual speech data annotation method was studied by identifying the corresponding language in the original data, and experiments were conducted on two public datasets. The results show that the proposed model achieves a substantial reduction in the average error rate 46.44% lower than the Conformer model, a mainstream speech recognition model, and 2.1% lower than the baseline model, thereby attaining higher accuracy in recognition. The research results provide novel perspectives and methodologies for the domain of cross-lingual speech recognition. [ABSTRACT FROM AUTHOR]

The acoustic input network based on the Conformer encoder has the problem of insufficient extraction of FBank speech information and missing channel feature information. This paper proposed an end-to-end method based on RepVGG-SE-Conformer for speech recognition to solve these problems. Firstly, the proposed model used the multi-branch structure of RepVGG to enhance the speech information extraction capability, and using the structural re-parameterization fused the multibranch into a single branch to reduce the computational complexity and speed up the model inference. Then, based on the squeeze-and-excitation network, the channel attention mechanism made up for the missing channel feature information to improve speech recognition accuracy. Finally, the experimental results on the public dataset Aishell-1 show that the proposed method's character error rate is reduced by 10.67% compared with Conformer, and the advancement of the method is verified. In addition, the proposed RepVGG-SE acoustic input network has good generalization ability in the end-to-end scene, which can effectively improve the overall performance of speech recognition models based on Transformer variants. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban) is the property of Zhongguo Xue shu qi Kan (Guang Pan Ban) Dian zi Za zhi She and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Articulation disorder, as a medical difficulty, currently mainstream speech recognition technologies are not well adapted to the needs of this field. At the same time, speech recognition technology for dysarthria utilizes a combination of pre training and personalized training to further improve algorithm performance and reduce recognition word error rate through data-driven methods. However, currently, speech recognition technology for dysarthria still has a certain distance from practical commercial use, and its development is limited by data scale and technology. So far, there have been no comprehensive articles on speech recognition for dysarthria. It is urgent to compare and analyze the construction methods and advanced technologies of various datasets in this field, in order to facilitate researchers entering the field to quickly acquire knowledge in this field. This paper conducts a survey on existing datasets, mainstream algorithms, and evaluation methods, and summarizes the scale, form, and characteristics of mainstream speech impairment datasets at home and abroad. It analyzes the mainstream algorithms for speech recognition with dysarthria, and provides the performance and characteristics of different algorithms. Finally, the performance evaluation indicators of the algorithm model based on the severity level of patients with dysarthria are studied, and future research directions are discussed, in order to provide help for the researchers engaged in speech recognition with dysarthria and assist in the rapid development of this field. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Shanghai University / Shanghai Daxue Xuebao is the property of Journal of Shanghai University (Natural Sciences) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Computer Measurement & Control is the property of Magazine Agency of Computer Measurement & Control and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban) is the property of Zhongguo Xue shu qi Kan (Guang Pan Ban) Dian zi Za zhi She and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

local feature information at shallow layers. To solve this problem, this study proposes a method using multiple encoders to improve the ability of speech feature extraction. An additional convolutional encoder branch is added to strengthen the capture of local feature information, make up for the neglect of local feature information in shallow Transformer, and effectively realize the integration of global and local dependencies of andio feature se quences. In other words, a multi-encoder model hased on Transformer is proposed. Experiments on the open source Chinese Mandarin data set Aishell 1 show that without an external language model, the proposed Transformer-based multi-encoder model has a relative reduction of 4.00% in character error rate when compared with the Transformer model. On the internal non public Shanghainese dialect data set, the performance improve ment of the proposed model is more obvious, and the character error rate is reduced by 48.24% from 19.92% to 10. 31%. [ABSTRACT FROM AUTHOR]

Copyright of Pearl River is the property of Pearl River Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Psychological Science is the property of Psychological Science Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban) is the property of Zhongguo Xue shu qi Kan (Guang Pan Ban) Dian zi Za zhi She and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)