本文介绍了语音识别系统芯片国产替代方案的实现。语音识别系统的硬件核心芯片采用富芮坤FR8016HA, 系统接收由移动端通过蓝牙发送来的音频信号, 之后对语音信号进行识别, 最终将识别结果呈现到LCD显示屏上, 实现人机交互。语音算法采用改进的DTW模板匹配法, 通过测试, 本系统语音识别正确率可达82%。
Abstract
In the paper, the realization of domestic alternatives for speech recognition system chips is introduced.The core chip of the speech recognition system is FR8016HA.The system receives the audio signal sent by the mobile terminal through Bluetooth, and then recognizes the voice signal, and finally presents the recognition result to the LCD display to realize man-machine interaction.The improved DTW template matching method is adopted in the speech algorithm.Through the test, the correct rate of speech recognition can reach 82%.
关键词
FR8016HA语音识别 /
DTW /
蓝牙接收
Key words
FR8016HA /
speech recognition /
DTW /
Bluetooth receiver
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 周红锴.基于单片机控制的孤立词语音自动识别系统设计[J].现代电子技术, 2020, 43(18):6466, 70.
[2] 姜琳, 周琳.国产芯片迎黄金发展期[J].中国品牌, 2020(9):8485.
[3] 方淼, 方帆.基于STM32的语音控制智能车实验系统[J].宿州学院学报, 2020, 35(1):6566, 70.
[4] 徐琳珊, 唐祁南, 姜林奇, 等.基于SPCE061A单片机的嵌入式语音识别系统的研究[J].电子技术, 2018, 47(12):3738.
[5] 陈盛, 胡维平, 张佑贤, 等.基于嵌入式的语音控制系统的设计与实现[J].电子设计工程, 2018, 6(19):5761.
[6] 叶硕, 彭春堂, 杜珍珍, 等.基于DTW的孤立词语音识别系统设计[J].长江大学学报(自然科学版), 2018, 15(17):3337, 5.
[7] 王晓冉, 曲立国, 方明星.基于STM32的人机语音交互系统[J].电子世界, 2017(13):104106.
[8] 吕修任.基于改进型DTW的语音识别算法设计与实现[D].南京:东南大学, 2017.
[9] 吴进, 张青.一种改进的孤立词语音识别系统设计[J].西安邮电大学学报, 2016, 21(1):7680.
[10] 吴进.语音信号处理实用教程[M].北京:人民邮电出版社, 2015.
[11] Huai Ximing, Kitada Satoshi, Choi Dongeun, et al.Heart sound recognition technology based on convolutional neural network[J].Informatics for Health and Social Care, 2021, 46(3).
[12] Jiang Hui, Qiao Qian, Zheng Hong, et al.Birdsong Recognition Based on Improved DTW[J].Journal of Physics: Conference Series, 2021, 1739(1).
