Voice data processing method and device, equipment, storage medium and program product

Abstract

The invention provides a voice data processing method and device, equipment, a storage medium and a program product. The method comprises the steps of obtaining collected to-be-processed voice data; obtaining an updated recognition model, wherein the updated recognition model is obtained by updating an initial recognition model by a terminal according to a personalized training data set, the initial recognition model is obtained by training through a server terminal based on public training data, and the personalized training data set at least comprises the collected voice data; inputting the to-be-processed voice data into the updated recognition model for recognition to obtain a recognition result; and determining a control instruction corresponding to the recognition result, and executing the control instruction. According to the invention, on the premise of ensuring that the user voice data is not out of the local and ensuring that the user privacy is not leaked, personalized voice recognition service with high recognition accuracy is realized.

Description

technical field [0001] This application relates to the technical field of artificial intelligence, and relates to but not limited to a voice data processing method, device, equipment, storage medium and program product. Background technique [0002] With the development of artificial intelligence, intelligent hardware and other fields, the human-computer interaction method based on speech recognition is more and more recognized by users. Especially in the vehicle scene, the driver can wake up the vehicle intelligent interactive system through voice, and then can issue control instructions through voice, which is convenient and safe. [0003] In the related art, the voice recognition service in the vehicle intelligent interactive system provides two methods: one is the cloud voice recognition service, the vehicle terminal uploads the user's voice to the cloud service provider server, and the voice recognition software on the cloud service provider server converts the voice to...

AI Technical Summary

Problems solved by technology

[0003] In the related art, the voice recognition service in the vehicle intelligent interactive system provides two methods: one is the cloud voice recognition service, the vehicle terminal uploads the user's voice to the cloud service provider server, and the voice recognition software on the cloud service provider server converts the voice to Translate it into text and send it back to the user. The advantage of the cloud service speech recognition service is that it has strong computing power and can continuously update the model. The voice recognition service is unavailable, and the user's voice needs to be uploaded. There is a risk of leaking the user's voiceprint information, and the security of private information such as the user's identity cannot be ensured; the other is to convert the user's voice into Text, the advantage is that it does not need to be connected to the Internet, and there is no privacy leakage problem, but it is limited by speech recognition technology, terminal computing and storage capabilities, complex command logic, complex background sounds, user accents and other factors, resulting in weak recognition ability and low recognition success rate. Low, and requires the cooperation of the user (including the use of standard Mandarin, louder voice, quiet background environment) to complete the recognition, which brings inconvenience to the user

Images