CN116778965A - Voice quality inspection device and system - Google Patents
Voice quality inspection device and system Download PDFInfo
- Publication number
- CN116778965A CN116778965A CN202310598066.4A CN202310598066A CN116778965A CN 116778965 A CN116778965 A CN 116778965A CN 202310598066 A CN202310598066 A CN 202310598066A CN 116778965 A CN116778965 A CN 116778965A
- Authority
- CN
- China
- Prior art keywords
- voice
- customer service
- quality inspection
- module
- layer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 31
- 238000011156 evaluation Methods 0.000 claims abstract description 9
- 230000001681 protective effect Effects 0.000 claims description 13
- 230000007246 mechanism Effects 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 5
- 238000013518 transcription Methods 0.000 claims description 5
- 230000035897 transcription Effects 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000012372 quality testing Methods 0.000 claims description 3
- 230000008451 emotion Effects 0.000 abstract description 3
- 230000009467 reduction Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000000034 method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/60—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for measuring the quality of voice signals
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/63—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for estimating an emotional state
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/42221—Conversation recording systems
Abstract
The invention discloses a device and a system for voice quality inspection, comprising: the application layer is used for carrying out actual application in multiple directions on the customer service record; the data processing layer is used for automatically identifying and processing the voice recordings of customer service; the storage layer is used for storing the voice records of customer service; the base equipment layer is matched through the use of various instruments. The invention utilizes the setting mode of matching the application layer and the data processing layer to carry out preliminary processing on the recording file through the audio processing module, such as denoising, noise reduction, signal enhancement and the like; by the identification and classification module: classifying and identifying the voice signals by utilizing acoustic and language models, such as voice recognition, voice emotion classification, voice gender classification, voice domain classification and the like; by means of an evaluation and feedback module: and evaluating the voice recognition result and feeding back the evaluation result to the application field so as to optimize the accuracy and efficiency of the application.
Description
Technical Field
The invention relates to the field of quality inspection devices, in particular to a device and a system for voice quality inspection.
Background
Currently, in the electric pin outbound service, manual spot check is the most common quality inspection means. The manual sampling inspection takes the qualification rate of part of samples as a delivery standard, and the traditional detection method is simple and visual, can be used as a standard for measuring business quality in most cases, but can not ensure the recall ratio when the number of individuals contained in the whole is large, and can also generate great labor cost and time cost; when the overall difference degree is large, the qualification rate representativeness obtained by the traditional sampling method is correspondingly reduced.
With the proposal of big data concept, the manual sampling inspection method exposes the problems of higher labor cost, higher time cost, lower recall ratio and the like, and the quality inspection of only call records obviously causes the waste of data.
Disclosure of Invention
The invention aims to provide a device and a system for voice quality inspection, which are used for solving the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions: a voice quality inspection system, comprising:
the application layer is used for carrying out actual application in multiple directions on the customer service record;
the data processing layer is used for automatically identifying and processing the voice recordings of customer service;
the storage layer is used for storing the voice records of customer service;
the base equipment layer is matched through the use of various instruments.
Preferably, the application layer comprises customer service recording quality inspection, voice information analysis, a service specification library and operation processing, wherein the customer service recording quality inspection is used for recording quality inspection of a customer speaking of customer service, the voice information analysis is used for analyzing voice information, the service specification library is used for storing rules such as service specification and the like, and the operation processing is used for analyzing and processing operation.
Preferably, the data processing layer includes a voice search service, a voice recognition analysis, a semantic understanding and a language model, the voice search service includes searching for a voice recording of a customer service, the voice recognition analysis is used for analyzing the voice recording of the customer service, the semantic understanding is used for carrying out semantic understanding analysis on the voice recording of the customer service, and the language model is used for establishing the language model.
Preferably, the storage layer comprises voice transcription, text analysis, recording backup and report generation, wherein the voice transcription is used for transcribing voice into text, the text analysis is used for analyzing the transcribed text, the recording backup is used for backing up the recording of customer service, and the report generation is used for automatically generating a report.
Preferably, the base device layer includes a server, a recording device, a network device, and a storage device.
Preferably, the customer service recording quality inspection comprises an audio processing module, a voice feature extraction module, a feature selection and preprocessing module, an acoustic model training module, a language model training module, an identification and classification module and an evaluation and feedback module.
An apparatus for voice quality testing, comprising: the recording device module comprises a neck wearing frame, earphone main bodies and a microphone mechanism, wherein the two earphone main bodies are respectively connected to two ends of the neck wearing frame, and one end of the microphone mechanism is connected with one earphone main body;
preferably, the microphone mechanism comprises an adjustable bracket, a microphone part, a protective sleeve and a connecting ring, wherein one end of the adjustable bracket is connected with the earphone main body, the other end of the adjustable bracket is connected with the microphone part, the protective sleeve is movably sleeved outside the microphone part, the connecting ring is fixedly connected with the protective sleeve, and the connecting ring is movably sleeved with the microphone part.
Preferably, the microphone part is connected with a plurality of limiting blocks in an annular array, the connecting ring is provided with a clamping groove, and the limiting blocks are mutually matched with the clamping groove.
The invention has the technical effects and advantages that:
the invention utilizes the setting mode of matching the application layer and the data processing layer to carry out preliminary processing on the recording file through the audio processing module, such as denoising, noise reduction, signal enhancement and the like; by the identification and classification module: classifying and identifying the voice signals by utilizing acoustic and language models, such as voice recognition, voice emotion classification, voice gender classification, voice domain classification and the like; by means of an evaluation and feedback module: and evaluating the voice recognition result and feeding back the evaluation result to the application field so as to optimize the accuracy and efficiency of the application.
Drawings
FIG. 1 is a data diagram of the system of the present invention.
FIG. 2 is a block diagram of a customer service recording quality inspection system according to the present invention.
Fig. 3 is a schematic structural diagram of the recording apparatus of the present invention.
Fig. 4 is a schematic diagram of a microphone mechanism according to the present invention.
Fig. 5 is an enlarged schematic view of the structure of fig. 4 a according to the present invention.
In the figure: 1. a neck-wearing frame; 2. an earphone body; 3. a microphone mechanism; 31. an adjustable bracket; 32. a microphone part; 321. a limiting block; 33. a protective sleeve; 34. and a connecting ring.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention provides a device and a system for voice quality inspection as shown in figures 1-5, wherein the voice quality inspection system comprises:
the application layer is used for carrying out actual application in multiple directions on the customer service record;
the data processing layer is used for automatically identifying and processing the voice recordings of customer service;
the storage layer is used for storing the voice records of customer service;
the base equipment layer is matched through the use of various instruments.
The application layer comprises customer service recording quality inspection, voice information analysis, a service standard library and operation processing, wherein the customer service recording quality inspection is used for recording quality inspection of a customer speaking call of a customer service, the voice information analysis is used for analyzing voice information, the service standard library is used for storing rules such as service standards, and the operation processing is used for analyzing and processing operation.
The data processing layer comprises a voice search service, voice recognition analysis, semantic understanding and a language model, wherein the voice search service comprises searching of customer service records, the voice recognition analysis is used for analyzing the customer service records, the semantic understanding is used for carrying out semantic understanding analysis on the customer service records, and the language model is used for establishing the language model.
The storage layer comprises voice transcription, text analysis, recording backup and report generation, wherein the voice transcription is used for transcribing voice into text, the text analysis is used for analyzing the transcribed text, the recording backup is used for backing up the recording of customer service, and the report generation is used for automatically generating a report.
The base device layer comprises a server, a recording device, a network device and a storage device.
The customer service recording quality inspection comprises an audio processing module, a voice feature extraction module, a feature selection and preprocessing module, an acoustic model training module, a language model training module, an identification and classification module and an evaluation and feedback module.
An apparatus for voice quality testing, comprising: the recording device module comprises a neck wearing frame 1, earphone main bodies 2 and a microphone mechanism 3, wherein the two earphone main bodies 2 are respectively connected to two ends of the neck wearing frame 1, and one end of the microphone mechanism 3 is connected with one earphone main body 2;
the microphone mechanism 3 comprises an adjustable bracket 31, a microphone part 32, a protective sleeve 33 and a connecting ring 34, wherein one end of the adjustable bracket 31 is connected with the earphone main body 2, the other end of the adjustable bracket 31 is connected with the microphone part 32, the protective sleeve 33 is movably sleeved outside the microphone part 32, the connecting ring 34 is fixedly connected with the protective sleeve 33, the connecting ring 34 is movably sleeved with the microphone part 32, the protective sleeve 33 is made of sponge, and the protection of the microphone part 32 is facilitated.
The microphone portion 32 is connected with a plurality of limiting blocks 321 in an annular array, the connecting ring 34 is provided with clamping grooves, and the limiting blocks 321 are matched with the clamping grooves, so that the position of the connecting ring 34 is conveniently positioned through the limiting of the limiting blocks 321, the connecting ring 34 is prevented from moving randomly, the protective sleeve 33 after being installed is conveniently positioned, and the protective sleeve 33 is prevented from moving randomly.
The audio processing module is used for carrying out preliminary processing on the record file, such as denoising, noise reduction, signal enhancement and the like; through the voice feature extraction module: extracting feature vectors in the voice signals from the processed audio so as to facilitate subsequent processing; through a feature selection and preprocessing module: the extracted feature vectors are selected and processed in a meaningful way, such as filtering and screening, or the feature extraction is carried out by utilizing a deep learning technology; training a module through an acoustic model: generating a corresponding model by fitting acoustic properties of the speech signal, such as speech phonemes, tones, inter-phoneme transition probabilities, etc.; through a language model training module: training the model according to the service scene to enable the model to have recognition capability for different languages, accents, contexts and fields; by the identification and classification module: classifying and identifying the voice signals by utilizing acoustic and language models, such as voice recognition, voice emotion classification, voice gender classification, voice domain classification and the like; by means of an evaluation and feedback module: and evaluating the voice recognition result and feeding back the evaluation result to the application field so as to optimize the accuracy and efficiency of the application.
Finally, it should be noted that: the foregoing description is only illustrative of the preferred embodiments of the present invention, and although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described, or equivalents may be substituted for elements thereof, and any modifications, equivalents, improvements or changes may be made without departing from the spirit and principles of the present invention.
Claims (9)
1. A voice quality testing system, comprising:
the application layer is used for carrying out actual application in multiple directions on the customer service record;
the data processing layer is used for automatically identifying and processing the voice recordings of customer service;
the storage layer is used for storing the voice records of customer service;
the base equipment layer is matched through the use of various instruments.
2. The voice quality inspection system according to claim 1, wherein the application layer comprises customer service recording quality inspection, voice information analysis, a service specification library and operation processing, the customer service recording quality inspection is used for recording quality inspection of customer speaking of customer service, the voice information analysis is used for analyzing voice information, the service specification library is used for storing rules such as service specification, and the operation processing is used for analyzing operation.
3. The voice quality inspection system of claim 1, wherein the data processing layer includes a voice search service that includes searching for customer service recordings, a voice recognition analysis for analyzing the customer service recordings, a semantic understanding for performing a semantic understanding analysis on the customer service recordings, and a language model for building the language model.
4. The voice quality inspection system of claim 1, wherein the storage layer comprises voice transcription for transcribing voice into text, text analysis for analyzing the transcribed text, recording backup for backing up a customer service recording, and report generation for automatically generating a report.
5. The voice quality inspection system of claim 1, wherein the base device layer comprises a server, a recording device, a network device, and a storage device.
6. The voice quality inspection system of claim 2, wherein the customer service record quality inspection comprises an audio processing module, a voice feature extraction module, a feature selection and preprocessing module, an acoustic model training module, a language model training module, a recognition and classification module, and an evaluation and feedback module.
7. The voice quality inspection device is characterized in that the recording equipment module comprises a neck wearing frame (1), an earphone main body (2) and a microphone mechanism (3), wherein the two earphone main bodies (2) are respectively connected to two ends of the neck wearing frame (1), and one end of the microphone mechanism (3) is connected with one of the earphone main bodies (2).
8. The device for voice quality inspection according to claim 7, wherein the microphone mechanism (3) comprises an adjustable bracket (31), a microphone portion (32), a protective sleeve (33) and a connecting ring (34), one end of the adjustable bracket (31) is connected with the earphone main body (2), the other end of the adjustable bracket (31) is connected with the microphone portion (32), the protective sleeve (33) is movably sleeved outside the microphone portion (32), the connecting ring (34) is fixedly connected with the protective sleeve (33), and the connecting ring (34) is movably sleeved with the microphone portion (32).
9. The device for detecting voice quality according to claim 8, wherein a plurality of limiting blocks (321) are connected to the microphone portion (32) in a ring-shaped array, a clamping groove is formed in the connecting ring (34), and the limiting blocks (321) are mutually matched with the clamping groove.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310598066.4A CN116778965A (en) | 2023-05-25 | 2023-05-25 | Voice quality inspection device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310598066.4A CN116778965A (en) | 2023-05-25 | 2023-05-25 | Voice quality inspection device and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116778965A true CN116778965A (en) | 2023-09-19 |
Family
ID=87988712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310598066.4A Pending CN116778965A (en) | 2023-05-25 | 2023-05-25 | Voice quality inspection device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116778965A (en) |
-
2023
- 2023-05-25 CN CN202310598066.4A patent/CN116778965A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112804400B (en) | Customer service call voice quality inspection method and device, electronic equipment and storage medium | |
| CN108074576B (en) | Speaker role separation method and system under interrogation scene | |
| CN111128223B (en) | Text information-based auxiliary speaker separation method and related device | |
| Ortega-Garcia et al. | AHUMADA: A large speech corpus in Spanish for speaker characterization and identification | |
| US8412530B2 (en) | Method and apparatus for detection of sentiment in automated transcriptions | |
| Muthusamy et al. | Reviewing automatic language identification | |
| CN107274916B (en) | Method and device for operating audio/video file based on voiceprint information | |
| US8078463B2 (en) | Method and apparatus for speaker spotting | |
| WO2020228173A1 (en) | Illegal speech detection method, apparatus and device and computer-readable storage medium | |
| WO2019148586A1 (en) | Method and device for speaker recognition during multi-person speech | |
| US20110004473A1 (en) | Apparatus and method for enhanced speech recognition | |
| CN112885332A (en) | Voice quality inspection method, system and storage medium | |
| CN111294468A (en) | Tone quality detection and analysis system for customer service center calling | |
| CN107886951B (en) | Voice detection method, device and equipment | |
| CN112951275B (en) | Voice quality inspection method and device, electronic equipment and medium | |
| Muthusamy et al. | Automatic language identification: a review/tutorial | |
| EP4078579A1 (en) | Emotion detection in audio interactions | |
| CN110689906A (en) | Law enforcement detection method and system based on voice processing technology | |
| CN110782902A (en) | Audio data determination method, apparatus, device and medium | |
| CN116246610A (en) | Conference record generation method and system based on multi-mode identification | |
| CN116778965A (en) | Voice quality inspection device and system | |
| KR102407055B1 (en) | Apparatus and method for measuring dialogue quality index through natural language processing after speech recognition | |
| CN109635151A (en) | Establish the method, apparatus and computer equipment of audio retrieval index | |
| CN115063155A (en) | Data labeling method and device, computer equipment and storage medium | |
| CN108364654A (en) | Method of speech processing, medium, device and computing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |