Supercharge Your Innovation With Domain-Expert AI Agents!

Recording control method and apparatus

A control method and voice data technology, applied in the field of data processing, can solve problems that affect user experience, affect the accuracy of voice recognition results, and cannot detect the end point of voice, so as to achieve the effect of improving user experience and accuracy

Active Publication Date: 2018-12-21
HISENSE
View PDF11 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, in practical applications, it is inevitable that the target speaker has stopped speaking, but there are other people around who start speaking suddenly. In such cases, the application of speech endpoint detection technology will not be able to detect the end point of the speech. This also causes the sound pickup device to continue recording after the target speaker has stopped speaking, which causes the sound pickup device to record undesired speech, which further affects the accuracy of subsequent speech recognition results based on the recorded speech and affects the user experience.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Recording control method and apparatus
  • Recording control method and apparatus
  • Recording control method and apparatus

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0070] See figure 2 , is a flow chart of an embodiment of a recording control method provided by an exemplary embodiment of this specification, and the method can be applied to figure 1 The example application scenario includes the following steps:

[0071] Step 201: Control the sound pickup device to start recording based on the specific voice data of the target speaker.

[0072] Combined with the above figure 1 For the relevant description of the application scenarios shown, in the embodiment of this specification, for the convenience of description, the above-mentioned activation words are referred to as specific voice data.

[0073] Step 202: Sampling the voice data recorded by the sound pickup device according to a preset sampling rule to obtain sampled data.

[0074] In the embodiment of this specification, after the sound pickup device 140 starts recording, first use the VAD detection algorithm to determine the voice start point, after the voice start point is deter...

Embodiment 2

[0094] See image 3 , is an embodiment flowchart of another recording control method provided by an exemplary embodiment of this specification, the image 3 The method exemplified in the above figure 1 the application scenarios shown, and figure 2 On the basis of the method shown, emphatically describe the process of training the speaker model, the method may include the following steps:

[0095] Step 301: Obtain at least one piece of historical voice data.

[0096] In the embodiment of this specification, during the daily use of the audio receiving device 140, the speech voices of all speakers can be collected, and the collected speech voices can be preprocessed, such as pre-emphasis, framing, windowing, etc., to obtain The voice data is saved, and for the convenience of description, the voice data is referred to as historical voice data.

[0097] In an embodiment, a collection period can be set, for example, one month from the first use of the radio device 140 is set as...

Embodiment 3

[0109] See Figure 4 , is an embodiment flowchart of another recording control method provided by an exemplary embodiment of this specification, the Figure 4 The method exemplified in the above figure 1 the application scenarios shown, and figure 2 On the basis of the shown method, emphatically describe the process of determining whether there is a target speaker model corresponding to the target speaker 110 in at least one speaker model obtained through training, the method may include the following steps:

[0110] Step 401: Perform feature extraction on the specific speech data of the target speaker to obtain feature values ​​of the specific speech data.

[0111] Step 402: For any speaker model in the at least one trained speaker model, perform operations on the speaker model and the feature values ​​of the specific speech data according to preset operation rules.

[0112] Step 403: Judging whether there is a speaker model whose calculation result is greater than the pr...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The application provides a recording control method and apparatus. The method comprises: when specific voice data of a target speaker are detected, a pickup device is controlled to start sound recording; according to a preset sampling rule, voice data recorded by the pickup device are sampled to obtain sampling data; feature extraction is carried out on the sampling data to obtain a feature value;the feature value is inputted into a trained target speaker model to obtain an output result of the target speaker model; and if the sampling data are determined not to match the target speaker basedon the output result, the pickup device is controlled to stop recording. With the method provided by the invention, the voice end point of the target speaker can be accurately determined and thus thevoice recorded by the pickup device only includes the speaking voice of the target speaker, so that the accuracy of voice identification based on the voice recorded by the pickup device is improved and thus the user experience is enhanced.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular to a recording control method and device. Background technique [0002] VAD (Voice Activity Detection, voice activity detection) detection algorithm is a voice endpoint detection technology, the application of this technology can accurately locate the start point and end point of the voice from the voice with noise. Based on this, in application scenarios involving recording, this technology can be applied to automatically control sound pickup devices, such as microphones, wearable devices, smart TVs, smart speakers, etc., to record when the target speaker is speaking. [0003] However, in practical applications, it is inevitable that the target speaker has stopped speaking, but there are other people around who start speaking suddenly. In such cases, the application of speech endpoint detection technology will not be able to detect the end point of the speech. Th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/04G10L15/06G10L17/00G10L17/04
CPCG10L15/04G10L15/06G10L17/00G10L17/04
Inventor 任晓楠崔保磊陈中帅
Owner HISENSE
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More