CN112562695B - Voiceprint recognition method, voiceprint recognition device, computer equipment and storage medium - Google Patents
Voiceprint recognition method, voiceprint recognition device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN112562695B CN112562695B CN202011348172.XA CN202011348172A CN112562695B CN 112562695 B CN112562695 B CN 112562695B CN 202011348172 A CN202011348172 A CN 202011348172A CN 112562695 B CN112562695 B CN 112562695B
- Authority
- CN
- China
- Prior art keywords
- voiceprint
- environmental
- voice
- home
- detecting whether
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000007613 environmental effect Effects 0.000 claims abstract description 157
- 238000012549 training Methods 0.000 claims description 34
- 238000004590 computer program Methods 0.000 claims description 26
- 238000013507 mapping Methods 0.000 claims description 15
- 238000001514 detection method Methods 0.000 claims description 11
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000012795 verification Methods 0.000 description 8
- 241000699666 Mus <mouse, genus> Species 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000004880 explosion Methods 0.000 description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 241001465754 Metazoa Species 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 241000238631 Hexapoda Species 0.000 description 1
- 241000699670 Mus sp. Species 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/26—Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/2803—Home automation networks
- H04L12/2816—Controlling appliance services of a home automation network by calling their functionalities
- H04L12/282—Controlling appliance services of a home automation network by calling their functionalities based on user interaction within the home
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Automation & Control Theory (AREA)
- Computer Networks & Wireless Communication (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The application provides a voiceprint recognition method, a voiceprint recognition device, computer equipment and a storage medium, wherein the method comprises the steps of detecting whether a trigger instruction of a mode leaving home is received or not; when receiving a trigger instruction of the leaving mode, entering the leaving mode, and detecting whether an environmental voiceprint exists; and when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information. Through triggering to enter the leaving mode, the voice recognition system can monitor indoor environment voiceprints in the leaving mode, so that safety monitoring of the home under the unmanned condition is realized, safety is effectively improved, and voiceprint recognition is more flexible.
Description
Technical Field
The present application relates to the field of voice recognition technologies, and in particular, to a method, an apparatus, a computer device, and a storage medium for voice print recognition.
Background
Voiceprint recognition is a technology for identity authentication based on voice, and belongs to a biological feature recognition technology. Voiceprint recognition has been widely used in smart homes, and the main application scenario is identity authentication of people or authority setting of some smart homes. Through voiceprint recognition, the intelligent system can recognize the voice of a user, and then control the household appliances or intelligent equipment in the house to work. However, the intelligent system can only identify voice prints of people at present, and has low flexibility.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a voiceprint recognition method, apparatus, computer device, and storage medium.
A voiceprint recognition method comprising:
detecting whether a trigger instruction of a mode leaving home is received;
when receiving a trigger instruction of the leaving mode, entering the leaving mode, and detecting whether an environmental voiceprint exists;
and when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information.
In one embodiment, when the environmental voiceprint is detected, the step of parsing the environmental voiceprint to obtain the current scene information includes:
when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain voiceprint information;
and obtaining the current scene information according to the voiceprint information and a preset voiceprint scene mapping relation.
In one embodiment, when the out-of-home mode trigger instruction is received, the step of entering the out-of-home mode and detecting whether the environmental voiceprint exists includes:
when a trigger instruction of the leaving mode is received, entering the leaving mode, detecting whether an environment voiceprint exists, and when a voice voiceprint is received, judging the voice voiceprint as the environment voiceprint.
In one embodiment, after entering the away-from-home mode, the step of determining the voice voiceprint as the environmental voiceprint when the voice voiceprint is received includes:
detecting whether the voice voiceprint carries a home mode triggering instruction or not when receiving the voice voiceprint;
when the voice voiceprint does not carry a home mode triggering instruction, judging the voice voiceprint as the environment voiceprint;
and when the voice voiceprint carries a home mode triggering instruction, entering a home mode, detecting whether the voice voiceprint is received or not, and stopping detecting the environment voiceprint.
In one embodiment, when the environmental voiceprint is detected, the step of parsing the environmental voiceprint to obtain the current scene information includes:
when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information, and recording the environmental voiceprint and the current scene information.
In one embodiment, the step of detecting whether the leave mode trigger instruction is received further includes:
acquiring environmental training voiceprints;
inputting the environmental training voiceprint into a residual error network for training to obtain an environmental voiceprint recognition model;
when the environmental voiceprint is detected, the step of analyzing the environmental voiceprint to obtain current scene information comprises the following steps:
and when the environmental voiceprint is detected, analyzing the environmental voiceprint based on the environmental voiceprint recognition model to obtain current scene information.
In one embodiment, when the out-of-home mode trigger instruction is received, the step of entering the out-of-home mode and detecting whether the environmental voiceprint exists further includes:
detecting whether a home mode trigger instruction is received or not;
and when receiving a home mode trigger instruction, entering a home mode, detecting whether voice voiceprint is received or not, and stopping detecting the environmental voiceprint.
A voiceprint recognition apparatus comprising:
the off-home mode trigger detection module is used for detecting whether an off-home mode trigger instruction is received or not;
the environment voiceprint detection module is used for entering a leaving mode when receiving a leaving mode trigger instruction and detecting whether environment voiceprints exist or not;
and the scene information analysis and acquisition module is used for analyzing the environment voiceprint to obtain current scene information when the environment voiceprint is detected.
A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor when executing the computer program performs the steps of:
detecting whether a trigger instruction of a mode leaving home is received;
when receiving a trigger instruction of the leaving mode, entering the leaving mode, and detecting whether an environmental voiceprint exists;
and when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
detecting whether a trigger instruction of a mode leaving home is received;
when receiving a trigger instruction of the leaving mode, entering the leaving mode, and detecting whether an environmental voiceprint exists;
and when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information.
According to the voiceprint recognition method, the voiceprint recognition device, the computer equipment and the storage medium, the voiceprint recognition system can monitor indoor environment voiceprints in the away-from-home mode by triggering the away-from-home mode, so that the safety of the home is monitored under the unmanned condition, the safety is effectively improved, and the voiceprint recognition is more flexible.
Drawings
FIG. 1 is a flow chart of a voiceprint recognition method in one embodiment;
FIG. 2 is a block diagram of a voiceprint recognition device in one embodiment;
FIG. 3 is an internal block diagram of a computer device in one embodiment;
FIG. 4 is a schematic diagram illustrating a voiceprint recognition method in one embodiment;
FIG. 5 is a schematic diagram of a process for constructing a voiceprint recognition model in one embodiment;
FIG. 6 is a schematic diagram of a training process for a voiceprint recognition model in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
Example 1
In this embodiment, as shown in fig. 1, a voiceprint recognition method is provided, which includes:
step 110, detecting whether a leave mode trigger instruction is received.
In this embodiment, the leave mode trigger instruction is used to trigger the intelligent voiceprint system or the voice recognition system in the room or home to enter the leave mode. The mode triggering instruction can be input by voice mode or by physical key.
In one embodiment, a voice voiceprint is detected, when the voice voiceprint is detected, the voice voiceprint is analyzed to obtain voice information, whether the voice information carries a leaving mode trigger instruction is detected, and when the voice information carries the leaving mode trigger instruction, a leaving mode is entered; and when the voice information is detected not to carry the off-home mode triggering instruction, executing control according to the voice information.
In one embodiment, whether a control instruction for a preset key is received is detected, and when the control instruction for the preset key is detected to be received, the reception of the leaving mode trigger instruction is determined.
And 120, entering a leaving mode when receiving a leaving mode trigger instruction, and detecting whether an environment voiceprint exists.
Specifically, the away mode is a working mode that the intelligent voiceprint system or the voice recognition system enters when a user in a home or a house leaves home and is not at home. In this away from home mode, the intelligent voiceprint system does not feed back voice voiceprints or voice information as no one is in home.
Specifically, the environmental voiceprint is a voiceprint of an environmental sound, and in one embodiment, the voiceprints corresponding to sounds other than the valid voice voiceprint are all environmental voiceprints. It should be understood that in this embodiment, the valid voice voiceprint refers to a voice voiceprint that can still perform voice control in the away mode, for example, a voice voiceprint that triggers release of the away mode.
In one embodiment, in the away mode, all of the voiceprints of the sound are ambient voiceprints. In one embodiment, in the away mode, the voiceprints of the sounds other than speech are all ambient voiceprints.
In this embodiment, when the leave mode trigger instruction is received, after entering the leave mode, no feedback is made to the voice voiceprint or the instruction of the voice voiceprint is executed, and the detection of the environmental voiceprint is shifted to. It should be appreciated that before entering the away-from-home mode, the intelligent voiceprint system or the voice recognition system is in the home mode, and in the home mode, no ambient voiceprints are detected, thus effectively reducing the consumption of computing resources.
And 130, when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information.
Specifically, in the off-home mode, when the environmental voiceprint is detected, the environmental voiceprint is analyzed to obtain current scene information of the house, and the current scene information of the house is used for reflecting the current scene of the house and the home. The scene reflects the current situation or state of the house. Therefore, the user away from home can know the current state of the house in time, and the safety of the user away from home is effectively improved.
In one embodiment, the sound corresponding to the environmental voiceprint includes a water stream sound, an animal sound, a door-knocked sound, an explosion sound, and a wind sound, for example, the animal sound includes a mouse sound, a pet sound, and an insect sound. The corresponding scene information comprises water leakage of families, mice in the families, unknown visitors, hidden explosion hazards, no door or window of the pets away from the families, and the like. Each voiceprint or a plurality of voiceprints are matched and correspond to scene information.
In the embodiment, the voice recognition system can monitor the indoor environment voiceprint in the away mode by triggering to enter the away mode, so that the safety of the home is monitored under the unmanned condition, the safety is effectively improved, and the voiceprint recognition is more flexible.
In order to obtain the scene information by parsing, in one embodiment, when the environmental voiceprint is detected, the step of parsing the environmental voiceprint to obtain the current scene information includes: when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain voiceprint information; and obtaining the current scene information according to the voiceprint information and a preset voiceprint scene mapping relation.
In this embodiment, the voiceprint scene mapping relationship is used to reflect the correspondence between the environmental voiceprint and the scene information, and it should be understood that, in one embodiment, the voiceprint scene mapping relationship is pre-constructed, and the voiceprint scene mapping relationship is obtained through a pre-constructed environmental voiceprint recognition model.
In one embodiment, each voiceprint or voiceprints cooperate to map a scene information, and the mapping relationship is as follows:
| identifying categories | Mapping scenes |
| Water flow sound | Water leakage exists in families |
| Mouse sound | Mouse in home |
| Door knock sound | With unknown guests |
| Sound of explosion | Has hidden danger of explosion |
| Sound of pet | The pet does not leave home |
| Sound of wind | Door and window is not closed |
In this way, the current scene information can be obtained by detecting the voiceprint information and finding the scene information corresponding to the voiceprint information in the voiceprint scene mapping relation.
In one embodiment, when the out-of-home mode trigger instruction is received, the step of entering the out-of-home mode and detecting whether the environmental voiceprint exists includes: when a trigger instruction of the leaving mode is received, entering the leaving mode, detecting whether an environment voiceprint exists, and when a voice voiceprint is received, judging the voice voiceprint as the environment voiceprint.
One embodiment is to enter a leave mode when receiving a leave mode trigger instruction, detect whether a voiceprint is received, detect whether the voiceprint is an ambient voiceprint or a voice voiceprint, identify the voice voiceprint as the ambient voiceprint when the voiceprint is a voice voiceprint, and not feedback the voice voiceprint.
In this embodiment, after entering the leaving mode, the received voice voiceprint is identified as an environmental voiceprint, and no feedback is made to the voice voiceprint. Therefore, in the away mode, voice voiceprints sent by users or other people cannot be identified and feedback is not carried out on the voice voiceprints, on one hand, consumption of computing resources is reduced, and on the other hand, environment voiceprints can be fully identified in the away mode, and interference of illegal or unexpected intruders on voiceprint identification by other people is avoided.
In one embodiment, after entering the away-from-home mode, the step of determining the voice voiceprint as the ambient voiceprint when the voice voiceprint is received includes: detecting whether the voice voiceprint carries a home mode triggering instruction or not when receiving the voice voiceprint; when the voice voiceprint does not carry a home mode triggering instruction, judging the voice voiceprint as the environment voiceprint; and when the voice voiceprint carries a home mode triggering instruction, entering a home mode, detecting whether the voice voiceprint is received or not, and stopping detecting the environment voiceprint.
In this embodiment, after entering the away mode, no feedback is made to the voice voiceprint or an instruction of the voice voiceprint is executed unless the voice voiceprint carries the home mode trigger instruction.
Specifically, the home mode triggering instruction is used for releasing the home mode, so that the intelligent voiceprint system or the voice recognition system releases the home mode and enters the home mode, and in the home mode, the intelligent voiceprint system or the voice recognition system does not monitor environmental voiceprints and only monitors voice voiceprints. In the embodiment, in the away mode, when a voice voiceprint is received, the voice voiceprint is analyzed to obtain voice information, whether the voice information carries a home mode triggering instruction is detected, when the voice information carries the home mode triggering instruction, the home mode is switched to, the voice voiceprint is monitored, and the monitoring of the environmental voiceprint is stopped; when the voice information does not carry the home mode triggering instruction, the home mode is continuously maintained, the environmental voiceprint is continuously monitored, and no feedback is made to the voice voiceprint.
In this embodiment, the intelligent voiceprint system or the voice recognition system only feeds back valid voiceprint information of the user in the away mode, where the valid voiceprint information is a voice voiceprint with a home mode trigger instruction sent by the user. On one hand, the method can effectively avoid interference of invalid voice voiceprints on voiceprint recognition in a home mode, can avoid feedback of the invalid voice voiceprints, effectively reduces consumption of computing resources, and on the other hand, can enable legal users to effectively and rapidly switch to the home mode.
In one embodiment, when the environmental voiceprint is detected, the step of parsing the environmental voiceprint to obtain the current scene information includes: when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information, and recording the environmental voiceprint and the current scene information.
In this embodiment, in the away-from-home mode, after the environmental voiceprint is parsed to obtain the current scene information, the environmental voiceprint and the current scene information are recorded, and the corresponding relationship between the environmental voiceprint and the current scene information is recorded, and the environmental voiceprint and the current scene information are sent to the user terminal, so that during the away-from-home period of the user, the impurities in the house can be recorded and timely sent to the user terminal, so that the user can learn in time.
In one embodiment, the step of detecting whether the leave mode trigger instruction is received further includes: acquiring environmental training voiceprints; inputting the environmental training voiceprint into a residual error network for training to obtain an environmental voiceprint recognition model; when the environmental voiceprint is detected, the step of analyzing the environmental voiceprint to obtain current scene information comprises the following steps: and when the environmental voiceprint is detected, analyzing the environmental voiceprint based on the environmental voiceprint recognition model to obtain current scene information.
In this embodiment, training is performed on various environmental voiceprints through a residual network, so as to identify types of environmental voiceprints, and in this embodiment, training and constructing an environmental voiceprint identification model includes three stages: training phase, registration phase and verification phase. Firstly, in a training stage, a large number of environmental training voiceprints are used as training data to be input into a residual error network for training, so that an environmental voiceprint recognition model is obtained, and the environmental voiceprint recognition model classifies different environmental training voiceprints; subsequently, in the registration stage, the registered environmental voiceprint is input into an environmental voiceprint recognition model to obtain registration features; and finally, in the verification stage, inputting any environmental voiceprint into the environmental voiceprint recognition model, extracting verification features through a residual network model, calculating the similarity between the verification features and registration features by using cosine, and judging the category of the environmental voiceprint according to the similarity. Thus completing the construction of the environmental voiceprint recognition model.
In one embodiment, when the out-of-home mode trigger instruction is received, the step of entering the out-of-home mode and detecting whether the environmental voiceprint exists further includes: detecting whether a home mode trigger instruction is received or not; and when receiving a home mode trigger instruction, entering a home mode, detecting whether voice voiceprint is received or not, and stopping detecting the environmental voiceprint.
In this embodiment, in the away-from-home mode, when a home mode trigger instruction is monitored, a home mode is entered, and in the home mode, a voice voiceprint is entered to be detected without detecting the environmental voiceprint, so that a user can conveniently control an intelligent home through voice when at home, and consumption of computing resources is reduced.
In one embodiment, after entering a home mode, or in the home mode, detecting voice voiceprints, analyzing the voice voiceprints to obtain voice information when the voice voiceprints are detected, detecting whether the voice information carries a home departure mode trigger instruction, and entering the home departure mode when the voice information carries the home departure mode trigger instruction; and when the voice information is detected not to carry the off-home mode triggering instruction, executing control according to the voice information. Thus, in the home mode, the user may control the intelligent speech recognition system and associated intelligent devices to operate via the voice voiceprint.
It should be understood that, although the steps in the flowchart of fig. 1 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 1 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of other steps or sub-steps of other steps.
Example two
In the current smart home leaving mode, the safety protection mainly depends on a camera, and the voice print recognition function is introduced here to increase the safety protection in the leaving scene. Voiceprint recognition is extended by the subject to other categories by the person. Thus, the safety protection of the smart home in the away mode can be enhanced. And some other sounds in the home are received into the voiceprint recognition system and are used for enhancing the safety protection of the home in the away-from-home scene.
The overall flow of the voiceprint recognition system is shown in fig. 4. The first step: when the user leaves home, starting a leave-home mode; and secondly, in the away mode, the voiceprint recognition module is adjusted, and the voiceprint recognition module is set as other voiceprint recognition modules by the main body. The subject of the previous voiceprint recognition is a person, either for authenticating the identity of the person or for making some settings of the device rights. The main body is replaced by other sounds and then used for protecting and early warning of household safety. The main intention of doing so here is to segment the scene, reduce the complexity of voiceprint recognition task and improve recognition accuracy; and thirdly, the voiceprint recognition system recognizes the corresponding category according to the monitored sound, and then maps the recognized category into a scene. Specific examples of the mapping are shown in the following table. And fourthly, writing the monitored time and event information into a log, and displaying the log on a terminal of the user in real time.
| Identifying categories | Mapping scenes |
| Water flow sound | Water leakage exists in families |
| Mouse sound | Mouse in home |
| Door knock sound | With unknown guests |
| Sound of explosion | Has hidden danger of explosion |
| Sound of pet | The pet does not leave home |
| Sound of wind | Door and window is not closed |
Next, the overall flow of voiceprint recognition will be described. The overall voiceprint recognition method is shown in fig. 5 and 6. The sounding class set in the voiceprint recognition system is not fixed, the fact that one class is added to retrain the model once is not realistic, and the residual network model plays a role in feature extraction in the whole method and is not used as a classifier or a recognizer. The method can be divided into three parts, training phase: training is a process of building a model, and training data is input into a residual error network for training; registration: the model is built in the training stage, and registered environment voiceprints are input into the model to obtain registered characteristics; verification: the verification stage is to input environmental voiceprints at will, extract verification features through a residual network model, and then calculate the similarity between the verification features and registration features by using cosine, and if the similarity is greater than a threshold value, the similarity is considered as a category.
Finally introducing the training of the residual error network model. At present, large data sets are all human voices, various voices in home are required to be collected manually, and the data size is not large. The network model is pre-trained on the public data set of the environmental voiceprint, then fine-tuned on the small data set, and if the user has special needs, the audio can be provided for personalized customization.
Example III
In this embodiment, as shown in fig. 2, a voiceprint recognition apparatus is provided, including:
the away-from-home mode trigger detection module 210 is configured to detect whether a away-home mode trigger instruction is received;
the environmental voiceprint detection module 220 is configured to enter a leaving mode when receiving a leaving mode trigger instruction, and detect whether an environmental voiceprint exists;
the scene information analysis obtaining module 230 is configured to analyze the environmental voiceprint to obtain current scene information when the environmental voiceprint is detected.
In one embodiment, the scene information parsing obtaining module includes:
the voiceprint information analysis and acquisition unit is used for analyzing the environmental voiceprint to acquire voiceprint information when the environmental voiceprint is detected;
and the scene mapping unit is used for obtaining the current scene information according to the voiceprint information and a preset voiceprint scene mapping relation.
In one embodiment, the environmental voiceprint detection module is configured to enter a leaving mode when receiving a leaving mode trigger instruction, detect whether an environmental voiceprint exists, and determine the voice voiceprint as the environmental voiceprint when receiving the voice voiceprint.
In one embodiment, the ambient voiceprint detection module includes:
the household trigger instruction detection unit is used for detecting whether the voice voiceprint carries a household mode trigger instruction or not when receiving the voice voiceprint;
the voice voiceprint judging unit is used for judging the voice voiceprint as the environment voiceprint when the voice voiceprint does not carry a home mode triggering instruction;
and the home mode triggering unit is used for entering a home mode when the voice voiceprint carries a home mode triggering instruction, detecting whether the voice voiceprint is received or not, and stopping detecting the environment voiceprint.
In one embodiment, the scene information analysis obtaining module is further configured to, when the environmental voiceprint is detected, analyze the environmental voiceprint to obtain current scene information, and record the environmental voiceprint and the current scene information.
In one embodiment, the voiceprint recognition apparatus further comprises:
the training voiceprint acquisition module is used for acquiring environment training voiceprints;
the environmental voiceprint recognition model construction module is used for inputting the environmental training voiceprint into a residual error network for training to obtain an environmental voiceprint recognition model;
and the scene information analysis and acquisition module is used for analyzing the environmental voiceprint based on the environmental voiceprint recognition model to acquire current scene information when the environmental voiceprint is detected.
In one embodiment, the voiceprint recognition apparatus further comprises:
the home instruction detection module is used for detecting whether a home mode trigger instruction is received or not;
and the home mode entering module is used for entering a home mode when receiving a home mode triggering instruction, detecting whether voice voiceprints are received or not, and stopping detecting the environment voiceprints.
For specific limitations of the voiceprint recognition device, reference may be made to the above limitations of the voiceprint recognition method, and no further description is given here. The individual units in the voiceprint recognition device described above can be implemented in whole or in part by software, hardware, and combinations thereof. The units can be embedded in hardware or independent of a processor in the computer equipment, and can also be stored in a memory in the computer equipment in a software mode, so that the processor can call and execute the operations corresponding to the units.
Example IV
In this embodiment, a computer device is provided. The internal structure thereof can be shown in fig. 3. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program, and the non-volatile storage medium is deployed with a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used to communicate with other computer devices in which application software is deployed. The computer program is executed by a processor to implement a voiceprint recognition method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in FIG. 3 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided comprising a memory storing a computer program and a processor that when executing the computer program performs the steps of:
detecting whether a trigger instruction of a mode leaving home is received;
when receiving a trigger instruction of the leaving mode, entering the leaving mode, and detecting whether an environmental voiceprint exists;
and when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information.
In one embodiment, the processor when executing the computer program further performs the steps of:
when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain voiceprint information;
and obtaining the current scene information according to the voiceprint information and a preset voiceprint scene mapping relation.
In one embodiment, the processor when executing the computer program further performs the steps of:
when a trigger instruction of the leaving mode is received, entering the leaving mode, detecting whether an environment voiceprint exists, and when a voice voiceprint is received, judging the voice voiceprint as the environment voiceprint.
In one embodiment, the processor when executing the computer program further performs the steps of:
detecting whether the voice voiceprint carries a home mode triggering instruction or not when receiving the voice voiceprint;
when the voice voiceprint does not carry a home mode triggering instruction, judging the voice voiceprint as the environment voiceprint;
and when the voice voiceprint carries a home mode triggering instruction, entering a home mode, detecting whether the voice voiceprint is received or not, and stopping detecting the environment voiceprint.
In one embodiment, the processor when executing the computer program further performs the steps of:
when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information, and recording the environmental voiceprint and the current scene information.
In one embodiment, the processor when executing the computer program further performs the steps of:
acquiring environmental training voiceprints;
inputting the environmental training voiceprint into a residual error network for training to obtain an environmental voiceprint recognition model;
and when the environmental voiceprint is detected, analyzing the environmental voiceprint based on the environmental voiceprint recognition model to obtain current scene information.
In one embodiment, the processor when executing the computer program further performs the steps of:
detecting whether a home mode trigger instruction is received or not;
and when receiving a home mode trigger instruction, entering a home mode, detecting whether voice voiceprint is received or not, and stopping detecting the environmental voiceprint.
Example five
In this embodiment, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
detecting whether a trigger instruction of a mode leaving home is received;
when receiving a trigger instruction of the leaving mode, entering the leaving mode, and detecting whether an environmental voiceprint exists;
and when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information.
In one embodiment, the computer program when executed by the processor further performs the steps of:
when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain voiceprint information;
and obtaining the current scene information according to the voiceprint information and a preset voiceprint scene mapping relation.
In one embodiment, the computer program when executed by the processor further performs the steps of:
when a trigger instruction of the leaving mode is received, entering the leaving mode, detecting whether an environment voiceprint exists, and when a voice voiceprint is received, judging the voice voiceprint as the environment voiceprint.
In one embodiment, the computer program when executed by the processor further performs the steps of:
detecting whether the voice voiceprint carries a home mode triggering instruction or not when receiving the voice voiceprint;
when the voice voiceprint does not carry a home mode triggering instruction, judging the voice voiceprint as the environment voiceprint;
and when the voice voiceprint carries a home mode triggering instruction, entering a home mode, detecting whether the voice voiceprint is received or not, and stopping detecting the environment voiceprint.
In one embodiment, the computer program when executed by the processor further performs the steps of:
when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information, and recording the environmental voiceprint and the current scene information.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring environmental training voiceprints;
inputting the environmental training voiceprint into a residual error network for training to obtain an environmental voiceprint recognition model;
and when the environmental voiceprint is detected, analyzing the environmental voiceprint based on the environmental voiceprint recognition model to obtain current scene information.
In one embodiment, the computer program when executed by the processor further performs the steps of:
detecting whether a home mode trigger instruction is received or not;
and when receiving a home mode trigger instruction, entering a home mode, detecting whether voice voiceprint is received or not, and stopping detecting the environmental voiceprint.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (8)
1. A method of voiceprint recognition comprising:
detecting whether a home mode trigger instruction is received, wherein the detecting whether the home mode trigger instruction is received comprises the following steps: detecting voice voiceprints, analyzing the voice voiceprints to obtain voice information when the voice voiceprints are detected, and detecting whether the voice information carries a leaving mode trigger instruction or not;
when receiving a home-leaving mode trigger instruction, entering a home-leaving mode, detecting whether an environment voiceprint exists, and when receiving a voice voiceprint, detecting whether the voice voiceprint carries a home-leaving mode trigger instruction; when the voice voiceprint does not carry a home mode triggering instruction, judging the voice voiceprint as the environment voiceprint; when the voice voiceprint carries a home mode triggering instruction, entering a home mode, detecting whether the voice voiceprint is received or not, and stopping detecting the environment voiceprint, wherein the voice voiceprint is used for executing voice control;
and when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information.
2. The method of claim 1, wherein when the environmental voiceprint is detected, the step of parsing the environmental voiceprint to obtain current scene information comprises:
when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain voiceprint information;
and obtaining the current scene information according to the voiceprint information and a preset voiceprint scene mapping relation.
3. The method of claim 1, wherein when the environmental voiceprint is detected, the step of parsing the environmental voiceprint to obtain current scene information comprises:
when the environmental voiceprint is detected, analyzing the environmental voiceprint to obtain current scene information, and recording the environmental voiceprint and the current scene information.
4. The method of claim 1, wherein the step of detecting whether an away mode trigger instruction is received further comprises:
acquiring environmental training voiceprints;
inputting the environmental training voiceprint into a residual error network for training to obtain an environmental voiceprint recognition model;
when the environmental voiceprint is detected, the step of analyzing the environmental voiceprint to obtain current scene information comprises the following steps:
and when the environmental voiceprint is detected, analyzing the environmental voiceprint based on the environmental voiceprint recognition model to obtain current scene information.
5. The method according to any one of claims 1-4, wherein the step of entering the away mode when receiving the away mode trigger command, and detecting whether the environmental voiceprint exists further comprises:
detecting whether a home mode trigger instruction is received or not;
and when receiving a home mode trigger instruction, entering a home mode, detecting whether voice voiceprint is received or not, and stopping detecting the environmental voiceprint.
6. A voiceprint recognition apparatus, comprising:
the off-home mode trigger detection module is used for detecting whether an off-home mode trigger instruction is received or not, and the detecting whether the off-home mode trigger instruction is received or not comprises the following steps: detecting voice voiceprints, analyzing the voice voiceprints to obtain voice information when the voice voiceprints are detected, and detecting whether the voice information carries a leaving mode trigger instruction or not;
the environment voiceprint detection module is used for entering a home departure mode when receiving a home departure mode trigger instruction, detecting whether an environment voiceprint exists or not, and detecting whether the voice voiceprint carries a home mode trigger instruction or not when receiving the voice voiceprint; when the voice voiceprint does not carry a home mode triggering instruction, judging the voice voiceprint as the environment voiceprint; when the voice voiceprint carries a home mode triggering instruction, entering a home mode, detecting whether the voice voiceprint is received or not, and stopping detecting the environment voiceprint;
and the scene information analysis and acquisition module is used for analyzing the environment voiceprint to obtain current scene information when the environment voiceprint is detected.
7. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 5 when the computer program is executed.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011348172.XA CN112562695B (en) | 2020-11-26 | 2020-11-26 | Voiceprint recognition method, voiceprint recognition device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011348172.XA CN112562695B (en) | 2020-11-26 | 2020-11-26 | Voiceprint recognition method, voiceprint recognition device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112562695A CN112562695A (en) | 2021-03-26 |
| CN112562695B true CN112562695B (en) | 2023-09-29 |
Family
ID=75046859
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011348172.XA Active CN112562695B (en) | 2020-11-26 | 2020-11-26 | Voiceprint recognition method, voiceprint recognition device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112562695B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW201416656A (en) * | 2012-10-16 | 2014-05-01 | Rong-Zhao Hong | Water leakage detection system of household water consumption equipment |
| CN103778916A (en) * | 2013-12-31 | 2014-05-07 | 三星电子（中国）研发中心 | Method and system for monitoring environmental sound |
| CN108200396A (en) * | 2018-01-05 | 2018-06-22 | 湖南固尔邦幕墙装饰股份有限公司 | Intelligent door system and intelligent door control method |
| CN108597164A (en) * | 2018-04-28 | 2018-09-28 | 上海与德科技有限公司 | A kind of theft preventing method, device, terminal and computer-readable medium |
| CN108877126A (en) * | 2017-05-12 | 2018-11-23 | 谷歌有限责任公司 | System, the method and apparatus of activity monitoring are carried out via house assistant |
| CN109074035A (en) * | 2016-04-22 | 2018-12-21 | 微软技术许可有限责任公司 | Multi-functional every room automated system |
| CN110189769A (en) * | 2019-05-23 | 2019-08-30 | 复钧智能科技(苏州)有限公司 | Abnormal sound detection method based on multiple convolutional neural networks models couplings |
| CN110393883A (en) * | 2019-08-28 | 2019-11-01 | 晋江市长胜电子科技有限公司 | Fire hydrant |
| CN110867959A (en) * | 2019-11-13 | 2020-03-06 | 上海迈内能源科技有限公司 | Intelligent monitoring system and monitoring method for electric power equipment based on voice recognition |
| CN111063359A (en) * | 2019-12-24 | 2020-04-24 | 太平金融科技服务（上海）有限公司 | Telephone return visit validity judging method, device, computer equipment and medium |
| CN111108554A (en) * | 2019-12-24 | 2020-05-05 | 广州国音智能科技有限公司 | Voiceprint recognition method based on voice noise reduction and related device |
| CN111508521A (en) * | 2019-01-30 | 2020-08-07 | 深圳市冠旭电子股份有限公司 | Security method, terminal device and storage medium |
-
2020
- 2020-11-26 CN CN202011348172.XA patent/CN112562695B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW201416656A (en) * | 2012-10-16 | 2014-05-01 | Rong-Zhao Hong | Water leakage detection system of household water consumption equipment |
| CN103778916A (en) * | 2013-12-31 | 2014-05-07 | 三星电子（中国）研发中心 | Method and system for monitoring environmental sound |
| CN109074035A (en) * | 2016-04-22 | 2018-12-21 | 微软技术许可有限责任公司 | Multi-functional every room automated system |
| CN108877126A (en) * | 2017-05-12 | 2018-11-23 | 谷歌有限责任公司 | System, the method and apparatus of activity monitoring are carried out via house assistant |
| CN108200396A (en) * | 2018-01-05 | 2018-06-22 | 湖南固尔邦幕墙装饰股份有限公司 | Intelligent door system and intelligent door control method |
| CN108597164A (en) * | 2018-04-28 | 2018-09-28 | 上海与德科技有限公司 | A kind of theft preventing method, device, terminal and computer-readable medium |
| CN111508521A (en) * | 2019-01-30 | 2020-08-07 | 深圳市冠旭电子股份有限公司 | Security method, terminal device and storage medium |
| CN110189769A (en) * | 2019-05-23 | 2019-08-30 | 复钧智能科技(苏州)有限公司 | Abnormal sound detection method based on multiple convolutional neural networks models couplings |
| CN110393883A (en) * | 2019-08-28 | 2019-11-01 | 晋江市长胜电子科技有限公司 | Fire hydrant |
| CN110867959A (en) * | 2019-11-13 | 2020-03-06 | 上海迈内能源科技有限公司 | Intelligent monitoring system and monitoring method for electric power equipment based on voice recognition |
| CN111063359A (en) * | 2019-12-24 | 2020-04-24 | 太平金融科技服务（上海）有限公司 | Telephone return visit validity judging method, device, computer equipment and medium |
| CN111108554A (en) * | 2019-12-24 | 2020-05-05 | 广州国音智能科技有限公司 | Voiceprint recognition method based on voice noise reduction and related device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112562695A (en) | 2021-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108305633B (en) | Speech verification method, apparatus, computer equipment and computer readable storage medium | |
| CN106251874B (en) | A kind of voice gate inhibition and quiet environment monitoring method and system | |
| WO2020140665A1 (en) | Method and apparatus for quality detection of double-recorded video, and computer device and storage medium | |
| CN107274916B (en) | Method and device for operating audio/video file based on voiceprint information | |
| Ntalampiras et al. | Probabilistic novelty detection for acoustic surveillance under real-world conditions | |
| CN112037791B (en) | Conference summary transcription method, apparatus and storage medium | |
| CN112991585B (en) | Access personnel management method and computer readable storage medium | |
| CN109036438A (en) | Storage cabinet control, device, computer equipment and storage medium | |
| CN109599117A (en) | A kind of audio data recognition methods and human voice anti-replay identifying system | |
| CN112562150A (en) | Student apartment management method, device, system and medium based on face recognition | |
| CN111063067A (en) | Intelligent access control system based on voice control | |
| CN103856614A (en) | Method and device for avoiding error hibernation of mobile terminal | |
| CN109829691B (en) | C/S card punching method and device based on position and deep learning multiple biological features | |
| CN104965589A (en) | Human living body detection method and device based on human brain intelligence and man-machine interaction | |
| TW202018577A (en) | Human recognition method based on data fusion | |
| CN110826484A (en) | Vehicle weight recognition method and device, computer equipment and model training method | |
| CN108540357B (en) | Voice control method and device and sound equipment | |
| CN112489276A (en) | Gate inhibition control method and device based on gait recognition and storage medium | |
| CN112562695B (en) | Voiceprint recognition method, voiceprint recognition device, computer equipment and storage medium | |
| CN114627581B (en) | Coerced fingerprint linkage alarm method and system for intelligent door lock | |
| CN113838469A (en) | Identity recognition method, system and storage medium | |
| CN113689613A (en) | Access control system, access control method, and storage medium | |
| CN109800727A (en) | A kind of monitoring method and device | |
| CN111276140B (en) | Voice command recognition method, device, system and storage medium | |
| CN214996794U (en) | Intelligent lock anti-theft device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |