CN115294986B - Method for reducing false awakening of intelligent voice interaction - Google Patents
Method for reducing false awakening of intelligent voice interaction Download PDFInfo
- Publication number
- CN115294986B CN115294986B CN202211228563.7A CN202211228563A CN115294986B CN 115294986 B CN115294986 B CN 115294986B CN 202211228563 A CN202211228563 A CN 202211228563A CN 115294986 B CN115294986 B CN 115294986B
- Authority
- CN
- China
- Prior art keywords
- data
- sound
- awakening
- voice interaction
- time
- 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
- 230000003993 interaction Effects 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000007405 data analysis Methods 0.000 claims abstract description 32
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 230000005540 biological transmission Effects 0.000 claims description 6
- 230000002452 interceptive effect Effects 0.000 abstract description 11
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
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
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0484—Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/4401—Bootstrapping
- G06F9/4418—Suspend and resume; Hibernate and awake
-
- 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
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Computer Security & Cryptography (AREA)
- Signal Processing (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention belongs to the technical field of voice interaction, and discloses a method for reducing false awakening of intelligent voice interaction, which comprises the following steps: the data analysis equipment receives x pieces of sound data y1 at the same moment, the sound data are collected by the intelligent voice interaction equipment, the recorded n pieces of sound fingerprint data are compared with the collected sound data y for analysis, if the recorded n pieces of sound fingerprint data are matched with the collected sound data y, the data analysis equipment is sent, the obtained sound data are marked as y1, and the sound fingerprint data comprise pre-stored sound frequency and pre-stored awakening words; establishing a data set for the x sound data y1, analyzing the data set, determining the awakening object, generating an awakening instruction after determining the awakening object, sending the awakening instruction to the intelligent voice interaction equipment corresponding to the awakening object, and awakening the corresponding intelligent voice interaction equipment according to the awakening instruction; let nearest intelligent voice interaction equipment and user interactive, interactive effect is better, can avoid many voice interaction equipment of same region to be awaken up simultaneously, effectively promotes mutual experience.
Description
Technical Field
The invention belongs to the technical field of voice interaction, and particularly relates to a method for reducing false awakening of intelligent voice interaction.
Background
With the popularization of intelligent electronic products and the development of voice technology, more and more electronic products can perform voice interactive operation by using voice control. For example, the intelligent sound box can be awakened by speaking a preset awakening word when in a dormant state, and receives a voice instruction sent by a user to perform voice interaction. The problem that exists at present is that a series of products of the same merchant set up the same awakening word, so in a scene with a plurality of products, any person saying the awakening word can awaken all devices in the awakening range, and the phenomenon of 'one-for-one-hundred-response' appears, which affects user experience.
In the prior art, a method for preventing false awakening is provided, and different awakening words are usually set for different intelligent sound boxes, but the memory burden of a user is undoubtedly increased, so that the intelligent sound box becomes the burden of the user.
In view of this, the inventor of the present application invented a method for reducing the false wake-up of intelligent voice interaction.
Disclosure of Invention
The invention aims to solve the technical problems and provides a method for reducing the false awakening of intelligent voice interaction.
The technical scheme of the invention is as follows: a method for reducing false awakening of intelligent voice interaction is applied to data analysis equipment and comprises the following steps:
the data analysis equipment receives x pieces of sound data y1 at the same moment, the sound data y are collected by the intelligent voice interaction equipment,
the sound data y comprises sound frequency and time for collecting the sound; comparing and analyzing the recorded n sound fingerprint data with the collected sound data y, if the n sound fingerprint data are matched with the collected sound data y, sending data analysis equipment, marking the obtained sound data as y1, if the n sound fingerprint data are not matched with the collected sound data y, not sending the data analysis equipment, wherein the value of n is an integer greater than 1, and the sound fingerprint data comprise pre-stored sound frequency and pre-stored awakening words; and x is an integer which is less than or equal to m and greater than 0, establishing a data set for the x sound data y1, analyzing the data set, determining the awakening object at this time, generating an awakening instruction after determining the awakening object, sending the awakening instruction to the intelligent voice interaction equipment corresponding to the awakening object, and awakening the corresponding intelligent voice interaction equipment according to the awakening instruction.
Preferably, in the above, the comparing and analyzing process of the collected voice data and the voice fingerprint data is as follows:
acquiring words in the collected sound data, comparing the sound frequency with a pre-stored sound frequency, comparing the words with pre-stored awakening words, and if the sound frequency is matched with the pre-stored sound frequency and the words are matched with the pre-stored awakening words, sending the obtained sound data y1 to data analysis equipment; otherwise, no transmission is performed.
Preferably, in the above, a data set is established for the x pieces of sound data y1, and the data set is analyzed to determine the current awakening object, where the determining process includes: and acquiring time t1 in the x pieces of sound data y1, and taking the intelligent voice interaction equipment corresponding to the minimum time t1 in the data set as a current awakening object.
Preferably, in the above, the sound data further includes a volume value z, if the number of the time t1 is multiple, the volume value z in the x sound data y1 is obtained, a sum of the time t1 and the volume value z corresponding to the time t1 is marked as an expression coefficient k, and the intelligent voice interaction device corresponding to the largest expression coefficient k1 in the data set is used as the present awakening object; and if the number of the expression coefficients k1 is multiple, randomly determining one expression coefficient k1, and taking the intelligent voice interaction equipment corresponding to the expression coefficient k1 as the awakening object at this time.
Preferably, in the above, before the collected sound data is compared and analyzed with the sound fingerprint data, the content of the words matched with the pre-stored wakeup word is marked as p1 and time s1, the content of the words matched with the pre-stored wakeup word is marked as p2 and time s2, the absolute value of s2 subtracted from s1 is marked as s, and whether to continue the comparison and analysis of the collected sound data and the sound fingerprint data is determined according to the magnitude of the absolute value s;
comparing and analyzing the magnitude of the absolute value s with a time threshold, and if the absolute value s is less than or equal to the time threshold, not continuing to compare and analyze the acquired sound data with the sound fingerprint data; and if the absolute value s is larger than the time threshold, continuously comparing and analyzing the acquired sound data and the sound fingerprint data.
A method for reducing false awakening of intelligent voice interaction is applied to intelligent voice interaction equipment and comprises the following steps:
the intelligent voice interaction equipment collects voice data y, wherein the voice data y comprises voice frequency and voice collection time; comparing and analyzing the recorded n sound fingerprint data with the collected sound data y, if the n sound fingerprint data are matched with the collected sound data y, sending data analysis equipment, marking the obtained sound data as y1, if the n sound fingerprint data are not matched with the collected sound data y, not sending the data analysis equipment, wherein the value of n is an integer greater than 1, and the sound fingerprint data comprise pre-stored sound frequency and pre-stored awakening words; the data analysis equipment receives x sound data y1 at the same time, x is an integer which is smaller than or equal to m and larger than 0, a data set is established for the x sound data y1, the data set is analyzed, the object awakened at this time is determined, an awakening instruction is generated after the awakening object is determined, the awakening instruction is sent to the intelligent voice interaction equipment corresponding to the awakening object, and the corresponding intelligent voice interaction equipment is awakened according to the awakening instruction.
Preferably, in the above, the comparing and analyzing process of the collected voice data and the voice fingerprint data is as follows:
acquiring words in the collected sound data, comparing the sound frequency with a pre-stored sound frequency, comparing the words with pre-stored awakening words, and if the sound frequency is matched with the pre-stored sound frequency and the words are matched with the pre-stored awakening words, sending the obtained sound data y1 to data analysis equipment; otherwise, no transmission is performed.
Preferably, in the above, a data set is established for the x pieces of sound data y1, and the data set is analyzed to determine the current awakening object, where the determining process includes: and acquiring time t1 in the x pieces of sound data y1, and taking the intelligent voice interaction equipment corresponding to the minimum time t1 in the data set as a current awakening object.
Preferably, in the above, the sound data further includes a volume value z, if the number of the time t1 is multiple, the volume value z in the x sound data y1 is obtained, a sum of the time t1 and the volume value z corresponding to the time t1 is marked as an expression coefficient k, and the intelligent voice interaction device corresponding to the largest expression coefficient k1 in the data set is used as the present awakening object; and if the number of the expression coefficients k1 is multiple, arbitrarily determining one expression coefficient k1, and taking the intelligent voice interaction equipment corresponding to the expression coefficient k1 as the awakening object of the time.
Preferably, in the above, before the comparison analysis of the collected sound data and the sound fingerprint data, the content of the word matching the pre-stored wakeup word is marked as p1 and time s1, the content of the word matching the pre-stored wakeup word is marked as p2 and time s2, the absolute value of s2 subtracted from s1 is marked as s, and whether the comparison analysis of the collected sound data and the sound fingerprint data is continued is determined according to the magnitude of the absolute value s;
comparing and analyzing the magnitude of the absolute value s with a time threshold, and if the absolute value s is less than or equal to the time threshold, not continuing to compare and analyze the acquired sound data with the sound fingerprint data; and if the absolute value s is larger than the time threshold, continuously comparing and analyzing the acquired sound data and the sound fingerprint data.
Due to the adoption of the technical scheme, the invention has the beneficial effects that:
(1) Through the sound data of the matching that a plurality of intelligence voice interaction equipment received, through confirming sound data time and sound size value, confirm the voice interaction equipment nearest from the sound source, let nearest intelligence voice interaction equipment interactive with the user, interactive effect is better, can avoid many voice interaction equipment of same region to be awaken up simultaneously, effectively promotes mutual experience.
(2) When a user uses the preset awakening word to talk about the voice interaction device, whether the voice interaction device is awakened or not is determined by pre-storing the appearance time of the awakening word and the appearance time of the content of the back part of the awakening word and calculating the difference value of the two times, and an awakening condition is added to the voice interaction device, so that the voice interaction device is prevented from being awakened by mistake in the situation, the intelligent degree of the voice interaction device is increased, and the use experience is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic diagram illustrating a method for reducing false wake-up of smart voice interaction according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
As shown in fig. 1, the method for reducing false wake-up of intelligent voice interaction according to this embodiment includes m intelligent voice interaction devices and a data analysis device, where the m intelligent voice interaction devices are in signal connection with the data analysis device.
The intelligent voice interaction equipment collects voice data y, wherein the voice data y comprises voice frequency and time for collecting the voice; the method comprises the steps of recording n sound fingerprint data, recording n value of an integer larger than 1 in advance by a user, wherein the sound fingerprint data comprise a pre-stored sound frequency and a pre-stored awakening word, and the pre-stored awakening word is a word for awakening the intelligent device, such as 'Xiaowang classmates'.
The intelligent voice interaction equipment compares and analyzes the acquired voice data y with the voice fingerprint data, if the acquired voice data y is matched with the voice fingerprint data y, the voice data y is sent to the data analysis equipment, the voice data obtained this time is marked as y1, and the voice data y is sent to the data analysis equipment; if not, not sending.
The data analysis equipment receives x sound data y1 at the same moment, wherein x is an integer smaller than or equal to m and larger than 0, a data set is established for the x sound data y1, the data set is analyzed, the awakening object at this time is determined, an awakening instruction is generated after the awakening object is determined, and the awakening instruction is sent to the intelligent voice interaction equipment corresponding to the awakening object.
Wherein, the comparison and analysis process of the collected sound data and the sound fingerprint data is as follows:
acquiring words in the collected sound data, comparing the sound frequency with a pre-stored sound frequency, comparing the words with pre-stored awakening words, and if the sound frequency is matched with the pre-stored sound frequency and the words are matched with the pre-stored awakening words, sending the obtained sound data y1 to data analysis equipment; otherwise, no transmission is performed.
The method comprises the following steps of establishing a data set for x sound data y1, analyzing the data set, and determining an awakening object at this time, wherein the determining step comprises the following steps: acquiring time t1 in x sound data y1, taking the intelligent voice interaction equipment corresponding to the minimum time t1 in a data set as a current awakening object, acquiring a volume value z in the x sound data y1 if the number of the time t1 is multiple, marking the sum of the time t1 and the volume value z corresponding to the time t1 as an expression coefficient k, and taking the intelligent voice interaction equipment corresponding to the maximum expression coefficient k1 in the data set as the current awakening object; and if the number of the expression coefficients k1 is multiple, arbitrarily determining one expression coefficient k1, and taking the intelligent voice interaction equipment corresponding to the expression coefficient k1 as the awakening object of the time.
Through the sound data of the matching that a plurality of intelligence pronunciation interactive device received, through confirming sound data time and sound magnitude value, confirm the pronunciation interactive device nearest from the sound source, let nearest intelligence pronunciation interactive device and user interactive, interactive effect is better, can avoid many pronunciation interactive devices in same region to be awaken up simultaneously, effectively promotes mutual experience.
In the voice interaction equipment at the in-service use in-process, often can replace voice interaction equipment itself with presetting awakening word, with family or friend discussion voice interaction equipment's use experience, promptly use "the queen classmate" to replace voice interaction equipment, when discussing "the queen classmate" with family or friend use experience, the "queen classmate" has appeared in this in-process sound data, but do not want to use "the queen classmate", voice interaction equipment is easily awaken up this moment, influence and use experience.
In order to solve the above situation, the design is further improved; before the collected sound data and the sound fingerprint data are compared and analyzed, the content matched with the pre-stored awakening word in the word is marked as p1 and time s1, the content of the part of the content matched with the pre-stored awakening word in the word is marked as p2 and time s2, the absolute value of s2 subtracted from s1 is marked as s, and whether the collected sound data and the sound fingerprint data are continuously compared and analyzed is judged according to the absolute value s.
Comparing and analyzing the absolute value s with a time threshold, wherein if the absolute value s is less than or equal to the time threshold, the time threshold can be set by a user or a developer, and the comparison and analysis of the acquired sound data and the sound fingerprint data is not continued; and if the absolute value s is larger than the time threshold, continuously comparing and analyzing the acquired sound data and the sound fingerprint data.
For example: the use experience of the Xiaowang classmates is good, the Xiaowang classmates in the sentence is p1, and the use experience of the Xiaowang classmates is good is p2.
When a user uses the preset awakening word to talk about the voice interaction device, whether the voice interaction device is awakened or not is determined by pre-storing the appearance time of the awakening word and the appearance time of the content of the back part of the awakening word and calculating the difference value of the two times, and an awakening condition is added to the voice interaction device, so that the voice interaction device is prevented from being awakened by mistake in the situation, the intelligent degree of the voice interaction device is increased, and the use experience is improved.
Example two
A method for reducing false awakening of intelligent voice interaction comprises m intelligent voice interaction devices and data analysis devices, wherein the m intelligent voice interaction devices are in signal connection with the data analysis devices.
The intelligent voice interaction device comprises a data acquisition module, a voice identity recording module and a verification module.
The data acquisition module acquires sound data y, and the sound data y comprises sound frequency and time for acquiring the sound.
The sound identity recording module records n sound fingerprint data, the value of n is an integer larger than 1, the n is recorded in advance by a user, the sound fingerprint data comprise pre-stored sound frequency and pre-stored awakening words, and the pre-stored awakening words are words for awakening the intelligent equipment, such as' classmates of King.
The verification module is used for comparing and analyzing the collected sound data y with the sound fingerprint data, sending the sound data y to the data analysis equipment if the collected sound data y is matched with the sound fingerprint data, marking the obtained sound data y as y1 and sending the marked sound data y to the data analysis equipment; and if not, not sending.
The data analysis equipment receives x sound data y1 at the same time, wherein x is an integer which is less than or equal to m and greater than 0, a data set is established for the x sound data y1, the data set is analyzed, the awakening object is determined, an awakening instruction is generated after the awakening object is determined, and the awakening instruction is sent to the intelligent voice interaction equipment corresponding to the awakening object.
Wherein, the comparison and analysis process of the collected voice data and the voice fingerprint data is as follows:
acquiring words in the collected sound data, comparing the sound frequency with a pre-stored sound frequency, comparing the words with pre-stored awakening words, and if the sound frequency is matched with the pre-stored sound frequency and the words are matched with the pre-stored awakening words, sending the obtained sound data y1 to data analysis equipment; otherwise, no transmission is performed.
The method comprises the following steps of establishing a data set for x sound data y1, analyzing the data set, and determining an awakening object at this time, wherein the determining step comprises the following steps: acquiring time t1 in x sound data y1, taking intelligent voice interaction equipment corresponding to the minimum time t1 in a data set as a current awakening object, acquiring volume values z in the x sound data y1 if the number of the time t1 is multiple, marking the sum of the time t1 and the volume values z corresponding to the time t1 as an expression coefficient k, and taking the intelligent voice interaction equipment corresponding to the maximum expression coefficient k1 in the data set as the current awakening object; and if the number of the expression coefficients k1 is multiple, randomly determining one expression coefficient k1, and taking the intelligent voice interaction equipment corresponding to the expression coefficient k1 as the awakening object at this time.
Before the collected sound data and the sound fingerprint data are compared and analyzed, the content matched with the pre-stored awakening word in the word is marked as p1 and time s1, the content of the part of the content matched with the pre-stored awakening word in the word is marked as p2 and time s2, the absolute value of s2 subtracted from s1 is marked as s, and whether the collected sound data and the sound fingerprint data are continuously compared and analyzed is judged according to the absolute value s.
Comparing and analyzing the absolute value s with a time threshold, wherein if the absolute value s is less than or equal to the time threshold, the time threshold can be set by a user or a developer, and the comparison and analysis of the acquired sound data and the sound fingerprint data is not continued; and if the absolute value s is larger than the time threshold, continuously comparing and analyzing the acquired sound data and the sound fingerprint data.
In the description herein, references to the description of "one embodiment," "an example," "a specific example," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (2)
1. A method for reducing false awakening of intelligent voice interaction is applied to data analysis equipment and is characterized by comprising the following steps:
the data analysis equipment receives x pieces of sound data y1 at the same moment, the sound data y is collected by the intelligent voice interaction equipment, and the sound data y comprises sound frequency and time for collecting the sound; comparing and analyzing the recorded n sound fingerprint data with the collected sound data y, if the n sound fingerprint data are matched with the collected sound data y, sending data analysis equipment, marking the obtained sound data as y1, if the n sound fingerprint data are not matched with the collected sound data y, not sending the data analysis equipment, wherein the value of n is an integer greater than 1, and the sound fingerprint data comprise pre-stored sound frequency and pre-stored awakening words; x is an integer which is less than or equal to m and greater than 0, a data set is established for x sound data y1, the data set is analyzed, the awakening object at this time is determined, an awakening instruction is generated after the awakening object is determined, the awakening instruction is sent to the intelligent voice interaction equipment corresponding to the awakening object, and the corresponding intelligent voice interaction equipment is awakened according to the awakening instruction;
the comparison and analysis process of the collected voice data and the voice fingerprint data is as follows:
acquiring words in the collected sound data, comparing the sound frequency with a pre-stored sound frequency, comparing the words with pre-stored awakening words, and if the sound frequency is matched with the pre-stored sound frequency and the words are matched with the pre-stored awakening words, sending the obtained sound data y1 to data analysis equipment; otherwise, the transmission is not carried out;
establishing a data set for the x sound data y1, analyzing the data set, and determining the awakening object, wherein the determination process comprises the following steps: acquiring time t1 in x pieces of sound data y1, and taking intelligent voice interaction equipment corresponding to the minimum time t1 in a data set as a current awakening object;
if the number of the time t1 is multiple, obtaining the volume values z in the x sound data y1, marking the sum of the volume values z corresponding to the time t1 and the time t1 as an expression coefficient k, and taking the intelligent voice interaction device corresponding to the maximum expression coefficient k1 in the data set as a current awakening object; if the number of the expression coefficients k1 is multiple, one expression coefficient k1 is determined arbitrarily, and the intelligent voice interaction device corresponding to the expression coefficient k1 is taken as the awakening object of the time;
before the collected sound data and the sound fingerprint data are compared and analyzed, marking the content matched with the pre-stored awakening word in the word as p1 and time s1, marking the part of the content matched with the pre-stored awakening word in the word as p2 and time s2, subtracting the absolute value of s2 from s1 as s, and judging whether to continue the comparison and analysis of the collected sound data and the sound fingerprint data or not according to the magnitude of the absolute value s;
comparing and analyzing the absolute value s with a time threshold, and if the absolute value s is less than or equal to the time threshold, not continuing to compare and analyze the acquired sound data with the sound fingerprint data; and if the absolute value s is larger than the time threshold, continuously comparing and analyzing the acquired sound data and the sound fingerprint data.
2. A method for reducing false awakening of intelligent voice interaction is applied to intelligent voice interaction equipment and is characterized by comprising the following steps:
the intelligent voice interaction equipment collects voice data y, wherein the voice data y comprises voice frequency and voice collection time; comparing and analyzing the recorded n sound fingerprint data with the acquired sound data y, if the recorded n sound fingerprint data are matched with the acquired sound data y, sending the data to a data analysis device, marking the obtained sound data as y1, and if the recorded n sound fingerprint data are not matched with the acquired sound data y, not sending the data to the data analysis device, wherein the value of n is an integer larger than 1, and the sound fingerprint data comprise pre-stored sound frequency and pre-stored awakening words; the data analysis equipment receives x sound data y1 at the same moment, wherein x is an integer which is less than or equal to m and is greater than 0, a data set is established for the x sound data y1, the data set is analyzed, a current awakening object is determined, an awakening instruction is generated after the awakening object is determined, the awakening instruction is sent to the intelligent voice interaction equipment corresponding to the awakening object, and the corresponding intelligent voice interaction equipment is awakened according to the awakening instruction;
the comparison and analysis process of the collected sound data and the sound fingerprint data is as follows:
acquiring words in the collected sound data, comparing the sound frequency with a pre-stored sound frequency, comparing the words with pre-stored awakening words, and if the sound frequency is matched with the pre-stored sound frequency and the words are matched with the pre-stored awakening words, sending the obtained sound data y1 to data analysis equipment; otherwise, the transmission is not carried out;
establishing a data set for the x sound data y1, analyzing the data set, and determining the awakening object, wherein the determination process comprises the following steps: acquiring time t1 in x pieces of sound data y1, and taking intelligent voice interaction equipment corresponding to the minimum time t1 in a data set as a current awakening object;
if the number of the time t1 is multiple, obtaining the volume values z in the x sound data y1, marking the sum of the volume values z corresponding to the time t1 and the time t1 as an expression coefficient k, and taking the intelligent voice interaction device corresponding to the maximum expression coefficient k1 in the data set as a current awakening object; if the number of the expression coefficients k1 is multiple, one expression coefficient k1 is determined arbitrarily, and the intelligent voice interaction device corresponding to the expression coefficient k1 is taken as the awakening object of the time;
before the collected sound data and the sound fingerprint data are compared and analyzed, marking the content matched with the pre-stored awakening word in the word as p1 and time s1, marking the part of the content matched with the pre-stored awakening word in the word as p2 and time s2, subtracting the absolute value of s2 from s1 as s, and judging whether to continue the comparison and analysis of the collected sound data and the sound fingerprint data or not according to the magnitude of the absolute value s;
comparing and analyzing the magnitude of the absolute value s with a time threshold, and if the absolute value s is less than or equal to the time threshold, not continuing to compare and analyze the acquired sound data with the sound fingerprint data; and if the absolute value s is larger than the time threshold, continuously comparing and analyzing the acquired sound data and the sound fingerprint data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211228563.7A CN115294986B (en) | 2022-10-09 | 2022-10-09 | Method for reducing false awakening of intelligent voice interaction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211228563.7A CN115294986B (en) | 2022-10-09 | 2022-10-09 | Method for reducing false awakening of intelligent voice interaction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115294986A CN115294986A (en) | 2022-11-04 |
| CN115294986B true CN115294986B (en) | 2023-01-10 |
Family
ID=83819266
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211228563.7A Active CN115294986B (en) | 2022-10-09 | 2022-10-09 | Method for reducing false awakening of intelligent voice interaction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115294986B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114860187A (en) * | 2018-01-03 | 2022-08-05 | 腾讯科技（深圳）有限公司 | Intelligent voice equipment control method and device, computer equipment and storage medium |
| CN111128150A (en) * | 2019-11-27 | 2020-05-08 | 云知声智能科技股份有限公司 | Method and device for awakening intelligent voice equipment |
| US20220215835A1 (en) * | 2021-01-06 | 2022-07-07 | Comcast Cable Communications, Llc | Evaluating user device activations |
| CN113345433B (en) * | 2021-05-30 | 2023-03-14 | 重庆长安汽车股份有限公司 | Voice interaction system outside vehicle |
| CN113948081A (en) * | 2021-10-09 | 2022-01-18 | 珠海格力电器股份有限公司 | Equipment awakening method and device, electronic equipment and storage medium |
-
2022
- 2022-10-09 CN CN202211228563.7A patent/CN115294986B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115294986A (en) | 2022-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109346061B (en) | Audio detection method, device and storage medium | |
| CN108039988B (en) | Equipment control processing method and device | |
| CN107147618A (en) | A kind of user registering method, device and electronic equipment | |
| CN111968644B (en) | Intelligent device awakening method and device and electronic device | |
| CN111814570B (en) | Face recognition method, system and storage medium based on dynamic threshold | |
| CN113076903A (en) | Target behavior detection method and system, computer equipment and machine readable medium | |
| CN110211609A (en) | A method of promoting speech recognition accuracy | |
| CN110766074B (en) | Method and device for testing identification qualification of abnormal grains in biological identification method | |
| CN111126147B (en) | Image processing method, device and electronic system | |
| CN111626229A (en) | Object management method, device, machine readable medium and equipment | |
| CN110111311B (en) | Image quality evaluation method and device | |
| CN115294986B (en) | Method for reducing false awakening of intelligent voice interaction | |
| CN106874852A (en) | A kind of device-fingerprint based on acceleration transducer is extracted and recognition methods | |
| CN107103269A (en) | One kind expression feedback method and intelligent robot | |
| CN113160815A (en) | Intelligent control method, device and equipment for voice awakening and storage medium | |
| CN107154996B (en) | Incoming call interception method and device, storage medium and terminal | |
| CN107832690B (en) | Face recognition method and related product | |
| CN115731620A (en) | Method for detecting counter attack and method for training counter attack detection model | |
| EP3828888B1 (en) | Method for recognizing at least one naturally emitted sound produced by a real-life sound source in an environment comprising at least one artificial sound source, corresponding apparatus, computer program product and computer-readable carrier medium | |
| CN114240926A (en) | Board card defect type identification method, device and equipment and readable storage medium | |
| CN113704544A (en) | Video classification method and device, electronic equipment and storage medium | |
| CN112633037A (en) | Object monitoring method and device, storage medium and electronic equipment | |
| CN113539284B (en) | Voice noise reduction method and device, computer equipment and storage medium | |
| Fox et al. | Robust multi-modal person identification with tolerance of facial expression | |
| CN113569770B (en) | Video detection method and device, electronic equipment and storage medium |
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 |