Check patentability & draft patents in minutes with Patsnap Eureka AI!

Telephone background music detection model method, system, device and medium

A background music and detection model technology, applied in telephone communication, speech analysis, electrical components, etc., can solve problems such as low work efficiency and wrong answers, and achieve the effects of improving work efficiency, improving generalization ability, and intelligent understanding and flexibility

Active Publication Date: 2021-06-11
CTRIP COMP TECH SHANGHAI
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] The technical problem to be solved by the present invention is to overcome the defects in the prior art that the OTA industry smart phone will produce wrong answers when facing the background music of the phone call and the manual customer service needs to wait when facing the background music of the phone advertisement, which leads to the problem of low work efficiency. Provide a telephone background music detection model method, system, equipment and medium

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
  • Telephone background music detection model method, system, device and medium
  • Telephone background music detection model method, system, device and medium
  • Telephone background music detection model method, system, device and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0084] This embodiment provides a phone background music detection model, such as figure 1 As shown, the telephone background music detection model includes an input layer 1, an audio CNN layer 2, a splicing layer 3, a convolutional layer 4, a fully connected layer 5, and an output layer 6.

[0085] The input layer 1 is used to receive the frame information of the phone background music, and perform feature extraction on the frame information to obtain features of preset dimensions.

[0086] The audio CNN layer 2 includes several parallel one-dimensional convolutional layers 21, the first pooling layer 22, and the Flatten layer 23. The audio CNN layer 2 is used to separately analyze the features of the preset dimensions through each one-dimensional convolutional layer 21. Perform convolution calculation and obtain the first feature data corresponding to each one-dimensional convolution layer 21; the first pooling layer 22 is connected to the tail of the one-dimensional convolu...

Embodiment 2

[0094] This embodiment provides a kind of telephone background music detection method, the telephone background music detection method application embodiment 1 in the telephone background music detection model is realized, as figure 2 Shown, the phone background music detection method includes:

[0095] Step 11, carry out mute cutting to input telephone voice and obtain audio segment;

[0096] Step 12, the audio segment is divided into frames to obtain corresponding frame information;

[0097] Step 13: Input the frame information into the telephone background music detection model to obtain the judgment result.

[0098] When the OTA intelligent customer service communicates with the guest or the hotel in real time through the phone, the guest or the hotel will have an automatic response ringtone, the hotel’s pre-recorded answer system recording, waiting beeps, hotel welcome words, and task buttons Process recordings, hotel advertisements, and English-sounding advertisements...

Embodiment 3

[0142] This embodiment provides a kind of telephone background music detection system, the telephone background music detection system application embodiment 1 in the telephone background music detection model is realized, as Image 6 As shown, the phone background music detection system includes a cutting module 201, a framing module 202, a preprocessing module 203, a training module 204 and a prediction module 205;

[0143] The cutting module 201 is used for silently cutting the input telephone voice to obtain audio segments.

[0144] The framing module 202 is used for framing the audio segment to obtain corresponding frame information.

[0145] When the OTA intelligent customer service communicates with the guest or the hotel in real time through the phone, the guest or the hotel will have an automatic response ringtone, the hotel’s pre-recorded answer system recording, waiting beeps, hotel welcome words, and task buttons Process recordings, hotel advertisements, and Engli...

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 invention discloses a telephone background music detection model, method, system, equipment and medium, wherein the input layer of the telephone background music detection model is used to receive the frame information of the telephone background music and perform feature extraction on the frame information to obtain the preset dimension Features; the audio CNN layer is used to convolve and calculate the features of the preset dimensions through each one-dimensional convolution layer and obtain the first feature data corresponding to each one-dimensional convolution layer; the splicing layer is used for the first feature The data is connected to obtain the second feature data; the multi-layer convolution layer is used to perform layer-by-layer convolution calculation on the second feature data to obtain the third feature data; the fully connected layer is used to obtain the probability of frame information according to the third feature data; output The layer is used to obtain the judgment result of whether it is the background music of the phone according to the probability. The telephone background music detection model can quickly and accurately detect the sound content of the telephone background music, and improves the generalization ability of the telephone background music detection model.

Description

technical field [0001] The invention relates to the field of speech algorithms, in particular to a method, system, device and medium for detecting a background music of a telephone. Background technique [0002] In the OTA (online travel) industry, when the OTA intelligent customer service or human customer service communicates with guests, enterprises or hotels in real time by telephone, if the guests, enterprises or hotels set up automatic response ringtones, or record in advance Good answer system recordings, waiting beeps, welcome words, task-type button process recordings, advertisements, and English pronunciation advertisements that contain sound content such as advertising background music, the intelligent customer service will understand the sound of the advertising background music as normal pronunciation The content may be regarded as a normal conversation and responded directly, resulting in wrong answers; while the human customer service needs to wait until the r...

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 Patents(China)
IPC IPC(8): H04M3/50H04M3/51G10L25/78G10L25/81
CPCG10L25/78G10L25/81H04M3/50H04M3/5175
Inventor 郝竹林罗超胡泓王俊彬
Owner CTRIP COMP TECH SHANGHAI
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