Keyword identification method and device and electronic equipment

Abstract

The invention relates to the technical field of electronic equipment, and discloses a keyword recognition method and device and electronic equipment, which are used for ensuring the balance between a recognition preparation rate and a false alarm rate. The method comprises the following steps: acquiring to-be-identified information, and decoding the to-be-identified information through a keyword set to obtain a first sequence; decoding the to-be-identified information through the full-amount set or the non-keyword set to obtain a second sequence; and determining the similarity between the first sequence and the second sequence, and outputting the first sequence as a keyword recognition result when the similarity meets a set threshold condition. According to the method, the first sequence is output as the keyword recognition result only when the similarity between the first sequence and the second sequence meets the set threshold condition, so that the balance between the keyword recognition accuracy and the false alarm rate can be ensured.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of speech recognition technology, and in particular, to a keyword recognition method, apparatus, and electronic device. Background Art

[0002] With the emergence of applications such as smart speakers and voice assistants, users use voice to communicate with machines. Speech keyword recognition is an important technology for realizing human-machine voice interaction and is widely used in various intelligent devices and speech retrieval systems to complete device wake-up and device control.

[0003] Currently, under noise-free or low-noise conditions, speech keyword recognition can achieve excellent application effects. However, in a real environment with complex background noise (for example, subway stations, KTVs, conference rooms, etc.), the performance will drop sharply (the false alarm rate rises sharply or the recognition accuracy drops sharply), and even cause the system to fail to work properly. Therefore, there is an urgent need for a recognition method that can ensure the performance of speech keyword recognition in different environments. Summary of the Invention

[0004] Embodiments of the present application provide a keyword recognition method, apparatus, and electronic device to ensure the balance between the accuracy rate and false alarm rate of keyword recognition.

[0005] In a first aspect, embodiments of the present application provide a keyword recognition method applied to an electronic device. In this method, the electronic device obtains information to be recognized; the electronic device decodes the information to be recognized through a keyword set to obtain a first sequence; the electronic device decodes the information to be recognized through a full set or a non-keyword set to obtain a second sequence; the electronic device determines the similarity between the first sequence and the second sequence, and when the similarity meets a set threshold condition, outputs the first sequence as the keyword recognition result. Optionally, the keyword set may include multiple keywords or may include a single keyword.

[0006] In this method, after obtaining the information to be recognized, the electronic device can decode the information to be recognized through the keyword set, the full set, or the non-keyword set respectively to obtain the first sequence and the second sequence, so that the electronic device can determine whether the similarity between the first sequence and the second sequence meets the set threshold condition to decide the final output of the keyword recognition result, achieving the balance between the recognition accuracy rate and the false alarm rate.

[0007] In a possible design, the electronic device may extract features from the information to be recognized to obtain the feature information of the information to be recognized; the electronic device may input the feature information into a keyword model to perform keyword recognition on the feature information based on the keyword model to obtain at least one first candidate sequence; the electronic device may output the first candidate sequence corresponding to the maximum likelihood probability as the first sequence.

[0008] With this design, the electronic device can output the first candidate sequence corresponding to the maximum likelihood probability identified by the keyword model as the first sequence, which can ensure the accuracy of keyword recognition.

[0009] In a possible design, the keyword model is constructed based on the first modeling unit, and the first modeling construction unit includes at least one of the pronunciation, phoneme, and phrase corresponding to the keyword set.

[0010] With this design, there are various construction methods for the keyword module, which can be applied to terminal devices of various scales.

[0011] In a possible design, the electronic device can extract features from the information to be recognized to obtain the feature information of the information to be recognized; the electronic device inputs the feature information into the background model and performs recognition on the feature information based on the background model to obtain at least one second candidate sequence; the electronic device can output the second candidate sequence corresponding to the maximum likelihood probability as the second sequence.

[0012] With this design, the electronic device can output the second candidate sequence corresponding to the maximum likelihood probability recognized by the background model as the second sequence, which can ensure the recognition accuracy.

[0013] In a possible design, the background model is constructed based on the second modeling unit, and the second modeling unit includes at least one of the pronunciation, phoneme, and phrase. Optionally, the second modeling unit includes at least one of the pronunciation, phoneme, and phrase corresponding to the full set or the non-keyword set.

[0014] With this design, the electronic device can construct the background model according to the storage resource and computing resource balance strategy. In this way, the background model can cover various forms of electronic devices, and the electronic device can select different background model construction methods according to different resource situations.

[0015] In a possible design, the electronic device can also determine the similarity according to the probability corresponding to the first sequence and the probability corresponding to the second sequence. Among them, the probability corresponding to the first sequence represents the probability of training the first sequence through the elements in the keyword set, and the probability corresponding to the second sequence represents the probability of training the second sequence through the elements in the full set or the non-keyword set.

[0016] With this design, the electronic device can determine the similarity between the first sequence and the second sequence by the probability of constructing the first sequence by the keyword model and the probability of constructing the second sequence by the background model, and can intuitively judge whether the keyword recognition is accurate.

[0017] In a second aspect, the present application provides a keyword recognition device, which includes:

[0018] An identification module is used to obtain information to be identified; decode the information to be identified through a keyword set to obtain a first sequence; decode the information to be identified through a full set or a non-keyword set to obtain a second sequence;

[0019] The determination module is used to determine the similarity between the first sequence and the second sequence; when the similarity meets a set threshold condition, the first sequence is output as a keyword recognition result.

[0020] In a third aspect, the present application provides an electronic device, comprising one or more processors and one or more memories. The one or more memories are used to store one or more computer programs and data information, and the one or more processors are used to execute the computer programs stored in the one or more memories, so that the electronic device performs the method described in any aspect of the first aspect. The electronic device may be the electronic device of the first aspect.

[0021] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program or instructions are stored. When the computer program or instructions are executed by a computing device, the computing device executes any aspect of the above-mentioned first aspect, and any possible implementation of any aspect.

[0022] In a fifth aspect, an embodiment of the present application provides a chip, which is coupled to a memory in an electronic device so that the chip calls a computer program stored in the memory during operation to implement a method that may be designed in any aspect of the first aspect of the embodiment of the present application.

[0023] In a sixth aspect, the present application provides a computer program product, which includes a computer program or instructions. When the computer program or instructions are executed by a computing device, the computing device executes the method in any aspect of the above-mentioned first aspect and any possible implementation of any aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a schematic diagram of a keyword recognition scenario;

[0025] Figure 2 A schematic diagram of a possible hardware structure of an electronic device provided as an example for this application;

[0026] Figure 3 A flowchart of a keyword identification method provided in this application;

[0027] Figure 4a A schematic diagram of a decoding search graph of a keyword space provided by the present application;

[0028] Figure 4b Schematic diagram of another decoding search graph of the keyword space provided by this application;

[0029] Figure 4c Schematic diagram of yet another decoding search graph of the keyword space provided by this application;

[0030] Figure 4d Schematic diagram of yet another decoding search graph of the keyword space provided by this application;

[0031] Figure 5 Schematic diagram of the decoding search graph of the full - scale decoding search space provided by this application;

[0032] Figure 6 Schematic diagram of the application architecture of an electronic device provided by this application;

[0033] Figure 7 Schematic diagram of the structure of a decoding graph module provided by this application;

[0034] Figure 8 Schematic diagram of the training of a decoding graph module provided by this application;

[0035] Figure 9 Schematic diagram of the complete process of a keyword recognition method provided by this application;

[0036] Figure 10 Schematic diagram of the structure of a keyword recognition device provided by this application. Detailed implementation manners

[0037] Next, the technical solutions in the embodiments of this application will be described in detail with reference to the accompanying drawings in the following embodiments of this application.

[0038] First, concepts related to the embodiments of this application will be explained.

[0039] (1) Keyword detection (KWS), which belongs to a sub - field in the field of speech recognition, aims to detect all occurrence positions of a specified word in a speech signal. Currently, it is mainly used for keyword detection and isolated word recognition in unconstrained speech. This application mainly uses it to complete isolated word recognition.

[0040] (2) Modeling unit refers to the basic unit for building a model.

[0041] (3) False alarm rate refers to the probability that, during the process of radar detection or other detections using the threshold detection method, due to the widespread existence and fluctuations of noise, a target is judged to exist when it actually does not. In this application, the false alarm rate refers to the probability that an electronic device judges that there is a keyword when there is actually no keyword.

[0042] In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after. "At least one (item)" or similar expressions refer to any combination of these items, including any combination of single item(s) or plural item(s). For example, at least one (item) of a, b, or c can represent: a, b, c, a and b, a and c, b and c, or a, b, and c, where a, b, and c can be single or multiple.

[0043] Moreover, unless otherwise stated, the ordinal numbers such as "first" and "second" mentioned in the embodiments of the present application are used to distinguish multiple objects, and are not used to limit the size, content, order, time sequence, priority, or importance of multiple objects. For example, the first file and the second file are only used to distinguish different files, rather than indicating differences in the size, content, priority, or importance of these two files.

[0044] Currently, under noise-free or low-noise conditions, voice keyword recognition can achieve excellent application results. However, in real environments with complex background noise (such as subway stations, KTVs, conference rooms, etc.), the performance will drop sharply (the false alarm rate rises sharply or the recognition accuracy drops sharply), and even cause the system to malfunction. Therefore, there is an urgent need for an identification method that can ensure the performance of voice keyword recognition in different environments.

[0045] To ensure the accuracy of keyword recognition, the embodiments of the present application provide a keyword recognition method applied to an electronic device. In this method, after obtaining the information to be recognized, the electronic device can decode the information to be recognized through a keyword set to obtain a first sequence, and decode the information to be recognized through a full set or a non-keyword set to obtain a second sequence. The electronic device can also determine the similarity between the first sequence and the second sequence, and when the similarity meets the set threshold condition, output the first sequence as the keyword recognition result. In this way, the electronic device can make a decision on the output of the keyword recognition result based on the obtained similarity, so as to ensure the accuracy of keyword recognition while making the false alarm rate of keyword recognition meet the actual requirements, and further ensure the balance between accuracy and false alarm rate.

[0046] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0047] The embodiments of the present application can be applied to various scenarios that perform related functions through keyword recognition. For example, asFigure 1 As shown, the user can communicate with the electronic device by voice. The electronic device collects voice in the surrounding environment to obtain the voice information to be recognized. The electronic device can perform keyword recognition on the information to be recognized. For another example, the electronic device can also be connected to an information collection device or a database to obtain the information to be recognized for keyword recognition.

[0048] In some examples, the electronic device can decode the voice information to be recognized through a keyword set to obtain a first sequence. At the same time, the electronic device can also decode the voice information to be recognized through a full set or a non-keyword set to obtain a second sequence. The electronic device can determine the output of the keyword recognition result according to the similarity between the first sequence and the second sequence. Exemplarily, when the similarity between the first sequence and the second sequence meets the set threshold condition, the electronic device can output the first sequence as the keyword recognition result. When the electronic device determines that the keyword recognition result is the first sequence, it can execute related functions, such as waking up the mobile phone.

[0049] Figure 2 The figure shows a schematic diagram of the hardware structure of a possible electronic device. The electronic device 100 can be the Figure 1 electronic device in. As Figure 2 shown, the electronic device 100 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone interface 170D, a sensor module 180, a button 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc. The sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, a barometric pressure sensor 180C, a magnetic sensor 180D, an acceleration sensor 180E, a distance sensor 180F, a proximity light sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, an ambient light sensor 180L, a bone conduction sensor 180M, etc.

[0050] The processor 110 may include one or more processing units. For example, the processor 110 may include an application processor (AP), a micro control unit (MCU), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Among them, different processing units may be independent devices or integrated in one or more processors. Among them, the controller may be the nerve center and command center of the electronic device 100. The controller can generate operation control signals according to the instruction operation code and timing signal to complete the control of fetching and executing instructions. A memory may also be provided in the processor 110 for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. This memory can save the instructions or data that the processor 110 has just used or recycled. If the processor 110 needs to use the instruction or data again, it can be directly called from the memory. This avoids repeated accesses, reduces the waiting time of the processor 110, and thus improves the efficiency of the system.

[0051] In the embodiments of the present application, the processor 110 may save one or more application programs, obtain information to be recognized, decode the information to be recognized through a keyword set to obtain a first sequence, and decode the information to be recognized through a full set or a non-keyword set to obtain a second sequence. In addition, the processor 110 may also determine the output of the keyword recognition result by determining whether the similarity between the first sequence and the second sequence meets the set threshold condition.

[0052] The USB interface 130 is an interface that conforms to the USB standard specification. Specifically, it may be a Mini USB interface, a Micro USB interface, a USB Type C interface, etc. The USB interface 130 can be used to connect a charger to charge the electronic device 100, or to transfer data between the electronic device 100 and peripheral devices. The charging management module 140 is used to receive charging input from the charger. The power management module 141 is used to connect the battery 142, the charging management module 140, and the processor 110. The power management module 141 receives the inputs from the battery 142 and / or the charging management module 140 to supply power to the processor 110, the internal memory 121, the external memory, the display screen 194, the camera 193, and the wireless communication module 160, etc.

[0053] The wireless communication function of the electronic device 100 can be implemented through Antenna 1, Antenna 2, the mobile communication module 150, the wireless communication module 160, the modulation and demodulation processor, and the baseband processor, etc. Antenna 1 and Antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in the electronic device 100 can be used to cover a single or multiple communication frequency bands. Different antennas can also be multiplexed to improve the utilization rate of the antennas. For example, Antenna 1 can be multiplexed as the diversity antenna of the wireless local area network. In some other embodiments, the antenna can be used in combination with a tuning switch.

[0054] The mobile communication module 150 can provide solutions for wireless communications such as 2G / 3G / 4G / 5G applied to the electronic device 100. The mobile communication module 150 can include at least one filter, switch, power amplifier, low noise amplifier (LNA), etc. The mobile communication module 150 can receive electromagnetic waves through Antenna 1, filter, amplify, etc. the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation. The mobile communication module 150 can also amplify the signal modulated by the modulation and demodulation processor, and convert it into electromagnetic waves through Antenna 1 and radiate it out. In some embodiments, at least some functional modules of the mobile communication module 150 can be arranged in the processor 110. In some embodiments, at least some functional modules of the mobile communication module 150 and at least some modules of the processor 110 can be arranged in the same device.

[0055] The wireless communication module 160 can provide solutions for wireless communications applied to the electronic device 100, including wireless local area network (WLAN) (such as Wi-Fi network), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared technology (IR), etc. The wireless communication module 160 can be one or more devices integrating at least one communication processing module. The wireless communication module 160 receives electromagnetic waves through Antenna 2, frequency-modulates and filters the electromagnetic wave signals, and sends the processed signals to the processor 110. The wireless communication module 160 can also receive the signals to be sent from the processor 110, frequency-modulate and amplify them, and convert them into electromagnetic waves through Antenna 2 and radiate them out.

[0056] In some embodiments, antenna 1 of electronic device 100 is coupled to mobile communication module 150, and antenna 2 is coupled to wireless communication module 160, such that electronic device 100 can communicate with a network and other devices through wireless communication technologies. The wireless communication technologies may include global system for mobile communications (GSM), general packet radio service (GPRS), code division multiple access (CDMA), wideband code division multiple access (WCDMA), time-division code division multiple access (TD-CDMA), long term evolution (LTE), the fifth generation (5G) mobile communication system, future communication systems such as the sixth generation (6G) system, etc., BT, GNSS, WLAN, NFC, FM, and / or IR technologies. The GNSS may include global positioning system (GPS), global navigation satellite system (GLONASS), beidou navigation satellite system (BDS), quasi-zenith satellite system (QZSS), and / or satellite based augmentation systems (SBAS).

[0057] The display screen 194 is used to display images, videos, etc. The display screen 194 includes a display panel. The display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode (AMOLED), a flexible light-emitting diode (FLED), a Miniled, a MicroLed, a Micro-oLed, a quantum dot light-emitting diode (QLED), etc. In some embodiments, the electronic device 100 may include one or N display screens 194, where N is a positive integer greater than 1. In the embodiments of the present application, the display screen 194 can be used to display the main interface, the application interface, the granularity adjustment control, etc.

[0058] The camera 193 is used to capture static images or videos. The camera 193 can include a front camera and a rear camera.

[0059] The internal memory 121 can be used to store computer-executable program code, and the executable program code includes instructions. The processor 110 executes various functional applications and data processing of the electronic device 100 by running the instructions stored in the internal memory 121. The internal memory 121 can include a program storage area and a data storage area. Among them, the program storage area can store the operating system and software codes of at least one application program (such as Huawei Video, Changlian, etc.). The data storage area can store the data generated during the use of the electronic device 100 (such as images, videos, etc.). In addition, the internal memory 121 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc.

[0060] The external memory interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the electronic device 100. The external memory card communicates with the processor 110 through the external memory interface 120 to achieve the data storage function. For example, files such as pictures and videos are saved in the external memory card.

[0061] The electronic device 100 can implement audio functions through an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone jack 170D, and an application processor, etc. For example, music playback, recording, etc.

[0062] It can be understood that Figure 2 the components shown do not constitute a specific limitation on the electronic device. The electronic device may also include more or fewer components than those shown, or combine certain components, or split certain components, or have different component arrangements. In the following embodiments, Figure 2 the electronic device 100 shown is taken as an example for introduction.

[0063] It should be noted that the electronic device involved in the keyword recognition method provided in the embodiments of the present application may be a terminal device or a server. Exemplarily, the terminal device may be a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, an augmented reality (AR) device, a virtual reality (VR) device, a laptop computer, a drone, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a smart speaker, a smart TV, and other intelligent devices. In the embodiments of the present application, no special limitation is imposed on the specific form of the electronic device. For example, the electronic device may be a mobile phone or a smart speaker.

[0064] The present application will be described below with reference to specific embodiments.

[0065] Figure 3 It is a schematic diagram of a keyword recognition method provided in an embodiment of the present application. As Figure 3 shown, the method includes:

[0066] S301: The electronic device obtains information to be recognized.

[0067] Among them, the information to be recognized may be any information that satisfies a finite search range. For example, the information to be recognized may be voice information, text information, blood pressure, heart rate. Among them, blood pressure and heart rate are waveform signals.

[0068] As an example, the electronic device may collect the surrounding environment to obtain the information to be recognized. For example, when the electronic device is a mobile phone and the information to be recognized is voice information, the mobile phone may automatically collect the surrounding environment to obtain the voice information.

[0069] As another example, the electronic device can also obtain the information to be recognized from an information collection device or a database that collects the information to be recognized. For example, when the information to be recognized is blood pressure, the user can use a sphygmomanometer to measure the blood pressure. The electronic device can be connected to the sphygmomanometer to obtain the blood pressure data measured by the sphygmomanometer.

[0070] S302: The electronic device decodes the information to be recognized through a keyword set to obtain a first sequence. The keyword set consists of one or more keywords.

[0071] In some examples, after obtaining the information to be recognized, the electronic device can extract features of the information to be recognized to obtain the feature information of the information to be recognized. Exemplarily, when the information to be recognized is voice information, the feature information can be mel-scale frequency cepstral coefficients (MFCC), etc. Another example is that when the information to be recognized is blood pressure data or heart rate, the electronic device can sample and quantize the blood pressure data or heart rate signal to convert the time-domain continuous signal into discrete data (array, matrix), and this discrete data is the feature information.

[0072] The electronic device can input the feature information of the information to be recognized into a keyword model, and perform keyword recognition on the feature information based on the keyword model to obtain at least one first candidate sequence. The keyword model is constructed based on the keyword set. The electronic device can output the first candidate sequence corresponding to the maximum likelihood probability as the first sequence. Exemplarily, when the keyword model performs keyword recognition on the feature information, outputting the first candidate sequence corresponding to the maximum likelihood probability as the recognition result of the keyword model can ensure the accuracy of the keyword model in keyword recognition.

[0073] Exemplarily, when the information to be recognized is voice information or text information, the keyword can be a word or a phrase, that is, the electronic device performs keyword recognition on the feature information based on the keyword model to determine the keyword sequence in the voice information or text information.

[0074] Another example is that when the information to be recognized is blood pressure data, the keyword can be a waveform signal with a mutation point, that is, the electronic device performs keyword recognition on the feature information of the blood pressure data based on the keyword model to determine the mutation value in the blood pressure data. The mutation value in the blood pressure data is the mutation point in the waveform signal of the blood pressure data.

[0075] Another example is that when the information to be recognized is heart rate data, the keyword can be a waveform signal with a mutation point, that is, the electronic device performs keyword recognition on the feature information of the heart rate data through the keyword model to identify the mutation value in the heart rate data.

[0076] The keyword model is constructed based on the first modeling unit, and the first modeling unit can be the basic constituent unit of the keyword set. In some embodiments, when the keyword is speech or text, the first modeling unit includes at least one of the corresponding pronunciation, phoneme, and word in the keyword set. Pronunciation, phoneme, and word are the basic constituent units of speech. For example, when the speech is Chinese speech, the pronunciation is the pinyin of a single Chinese character, the phoneme is the initial and final consonants, and the word is a Chinese character phrase. Exemplarily, the electronic device can construct a decoding search graph of the keyword space based on pronunciation, phoneme, or word, and use the constructed decoding search graph of the keyword space as the keyword model. Additionally, multiple groups of keywords can be constructed in the keyword model.

[0077] As an example, the keyword model can be constructed based on the pronunciation corresponding to the keyword set. For example, as Figure 4a shown, when the keyword set includes the keyword "Hello, Xiaoka", the corresponding pronunciations of the keyword set are "ni", "hao", "xiao", "ka". The electronic device performs keyword recognition on the feature information based on the keyword model, obtains one or two first candidate sequences, and calculates the likelihood probability corresponding to the first candidate sequence. Among them, the first candidate sequence can be "ni", "hao", "xiao", "ka", or can be an empty sequence "non". Among them, the keyword model takes all sequences different from the keyword sequence "ni", "hao", "xiao", "ka" as garbage edges, that is, empty sequences. When the first candidate sequence corresponding to the maximum likelihood probability is "ni", "hao", "ni", "hao", the electronic device outputs "ni", "hao", "ni", "hao" as the first sequence.

[0078] Another example, as Figure 4b shown, when the keyword set includes two groups of keywords "Hello, Xiaoka" and "Hello, Hello", the keyword model has two sequences of pronunciations corresponding to the keywords "ni", "hao", "xiao", "ka" and "ni", "hao", "ni", "hao", and one garbage edge. The electronic device performs keyword recognition on the feature information based on the keyword model, obtains at least one first candidate sequence, and calculates the likelihood probability corresponding to at least one first candidate sequence respectively. Among them, the first candidate sequence can be "ni", "hao", "xiao", "ka", can also be "ni", "hao", "ni", "hao", or can be an empty sequence. When the first candidate sequence corresponding to the maximum likelihood probability is "ni", "hao", "ni", "hao", the electronic device outputs "ni", "hao", "ni", "hao" as the first sequence.

[0079] As another example, the keyword model can also be constructed based on the phonemes corresponding to the keyword set. For example, as Figure 4c shown, when the keyword set includes a single keyword "Hello", the phonemes corresponding to the keyword set are "n", "i", "h", "ao", with a total of 4 phoneme elements. This application can construct a decoding search graph for the keyword space based on these 4 phoneme elements.

[0080] As another example, the keyword model can also be constructed based on the phrases corresponding to the keyword set. For example, when the keyword set includes the keyword "Hello, Xiaoka", the phrase corresponding to the keyword set consists of "Ni", "Hao", "Xiao", "Ka". As Figure 4d shown, this application can construct a decoding search graph for the keyword space based on the keyword phonetic sounds, where the decoding search graph has a total of 4 phonetic sound elements.

[0081] In some other embodiments, when the information to be recognized is blood pressure data or heart rate, the first modeling unit may include information such as the waveform amplitude, peak, and trough corresponding to the mutation point. Exemplarily, when the information to be recognized is blood pressure data, the electronic device can construct a keyword recognition model based on the waveform amplitude corresponding to the mutation point. After the electronic device inputs the characteristic information of the blood pressure data into the keyword model, it performs keyword recognition based on the keyword model to determine at least one first candidate sequence. The first candidate sequence can be an array with a mutation point or an empty sequence. Among them, the keyword model sets all arrays without mutation points as empty sequences. The electronic device can respectively determine the likelihood probabilities corresponding to at least one first candidate sequence and output the first candidate sequence corresponding to the maximum likelihood probability as the first sequence.

[0082] S303: The electronic device decodes the information to be recognized through the full set or the non-keyword set to obtain a second sequence. The full set includes the keyword set and other information, and the non-keyword set includes the information in the full set except the keyword set. For example, when the keyword set consists of "Hao", the full set consists of all Chinese characters, and the non-keyword set consists of Chinese characters other than "Hao".

[0083] In some embodiments, before the electronic device decodes the information to be recognized through the full set or the non-keyword set, it can also extract the features of the information to be recognized to obtain the characteristic information of the information to be recognized. The process by which the electronic device obtains the characteristic information is the same as the process of determining the characteristic information in step S302.

[0084] The electronic device can also input the feature information into a background model (BGM), decode the feature information based on the background model, and obtain at least one second candidate sequence. Among them, the background model is constructed based on the full set or the non-keyword set. The electronic device can output the second candidate sequence with the maximum likelihood probability as the second sequence. Among them, the background model in the embodiments of the present application can absorb non-keyword information or noise input to reduce environmental interference.

[0085] The background model is constructed based on a second modeling unit, and the second modeling unit can be a basic constituent unit of the full set or the non-keyword set. In some embodiments, when the keyword is speech or text, the second modeling unit includes at least one of the corresponding Chinese characters, phonemes, and phrases of the full set or the non-keyword set. Exemplarily, the second modeling unit can include at least one of the full set of Chinese characters, the full set of phonemes, and the full set of phrases, or can also include the Chinese characters, phonemes, and phrases corresponding to the non-keyword.

[0086] The present application can construct a decoding search graph for the full decoding search space or the non-keyword space based on Chinese characters, phonemes, or phrases, and use the constructed decoding search graph for the full decoding search space or the non-keyword space as the background model. Exemplarily, the present application can select different second modeling units to construct the background model based on the resource situation of the electronic device. Further, the present application can select a suitable modeling unit to construct the background model in combination with the storage resource and computing resource balancing strategy of the electronic device.

[0087] As an example, when the computing resources are sufficient and the storage resources are tight, the present application can adopt a model construction method based on phonemes. Exemplarily, the present application can construct a decoding search graph for the background space based on the full set of phonemes or non-keyword phonemes to obtain the background model. For example, when the information to be recognized is Chinese speech, the present application can construct a decoding search graph for the background space based on the full set of phonemes. Among them, the full set of phonemes includes Chinese initials and finals, initials: b, p, m, f, d, t, n, l, g, k, h, j, q, x, zh, ch, sh, z, c, s, y, w, r;

[0088] ong.

[0090] As another example, when both computing resources and storage resources are abundant, the present application can adopt a model construction method based on phonetic sounds. Exemplarily, the present application can construct a decoding search graph for the background space based on all phonetic sounds or non-keyword phonetic sounds to obtain a background model. For example, when the information to be recognized is Chinese and the keyword is "Hello, Xiaoka", the present application can construct a decoding search graph for the background space based on all phonetic sounds. Among them, the decoding search graph of the background space contains 4 phonetic sound elements of the keyword "Hello, Xiaoka", namely "ni", "hao", "xiao", and "ka". Another example is that when the present application constructs a decoding search graph for the background space based on non-keyword phonetic sounds, the decoding search graph does not contain these 4 phonetic sound elements of "ni", "hao", "xiao", and "ka".

[0091] As another example, when computing resources are scarce and storage resources are sufficient, the present application can adopt a model construction method based on phrases. Exemplarily, the present application can construct a decoding search graph for the background space based on all phonetic sounds or non-keyword phonetic sounds to obtain a background model. For example, as Figure 5 shown, when the information to be recognized is Chinese and the keyword is "Hello, Xiaoka", the present application can construct a decoding search graph for the background space based on all phrases. Among them, Figure 5 the shown decoding search graph of the background space contains 4 phrase elements of the keyword "Hello, Xiaoka", namely "ni", "hao", "xiao", and "ka". Another example is that when the information to be recognized is Chinese and the keyword is "Hello, Xiaoka", the present application can also construct a decoding search graph for the background space based on non-keyword phrases. Among them, the decoding search graph of this background space does not contain 4 phrase elements of the keyword "Hello, Xiaoka", namely "ni", "hao", "xiao", and "ka".

[0092] In some other embodiments, when the information to be recognized is blood pressure data or heart rate, the second modeling unit may include information such as waveform amplitude, wave peak, and wave trough, which is not limited herein. Exemplarily, when the information to be recognized is blood pressure data, the electronic device can construct a background model based on the waveform amplitude. After the electronic device inputs the characteristic information of the blood pressure data into the background model, it performs recognition based on the background model to determine at least one second candidate sequence. Among them, the second candidate sequence is a waveform array. The electronic device can respectively determine the likelihood probabilities corresponding to at least one second candidate sequence, and output the second candidate sequence corresponding to the maximum likelihood probability as the second sequence.

[0093] S304: The electronic device determines the similarity between the first sequence and the second sequence.

[0094] In some embodiments, after the electronic device outputs the first sequence through the keyword model and the second sequence through the background model, it can also determine the similarity between the first sequence and the second sequence according to the probability corresponding to the first sequence and the probability corresponding to the second sequence. Among them, the probability corresponding to the first sequence represents the probability of training the first sequence through the elements in the keyword set, and the probability corresponding to the second sequence represents the probability of training the second sequence through the elements in the full set or the non-keyword set. Exemplarily, the closer the probability corresponding to the first sequence is to the probability corresponding to the second sequence, the higher the similarity between the first sequence and the second sequence.

[0095] As an example, the probability corresponding to the first sequence and the probability corresponding to the second sequence can be likelihood probability or posterior probability. In addition, the probabilities corresponding to the first sequence and the second sequence can also be other probabilities, which are not limited herein.

[0096] As an example, the electronic device can use the ratio of the probability corresponding to the first sequence to the probability corresponding to the second sequence as the similarity between the first sequence and the second sequence. Among them, the probability corresponding to the first sequence is the probability that the elements in the keyword model form the first sequence, and the probability corresponding to the second sequence is the probability that the elements in the background model form the second sequence. In addition, the electronic device can also determine the similarity between the first sequence and the second sequence through other calculation methods, which are not limited herein.

[0097] Exemplarily, the embodiments of the present application can determine the similarity between the first sequence and the second sequence through the following formula: R likelihood =L kws / L bg ; where R likelihood is the similarity, L kws represents the probability corresponding to the first sequence output by the keyword model, and L bg represents the probability corresponding to the second sequence output by the background model. The electronic device can use R likelihood as the similarity between the first sequence and the second sequence. Among them, the closer R likelihood is to 1, the higher the similarity between the first sequence and the second sequence. For example, when L kws represents the likelihood probability corresponding to the first sequence, and L bg represents the likelihood probability of the second sequence, the similarity R likelihood is the likelihood probability ratio.

[0098] For example, as Figure 4d shown, each arrow in the decoding search graph of the keyword model has its corresponding weight value. Among them, the weight value represents the probability that the keyword model selects this element. The probability corresponding to the first sequence is the probability L that the 4 elements of "you", "hao", "xiao", and "ka" in the keyword model form the path of "you"-"hao"-"xiao"-"ka"kws = m1 * m2 * m3 * m4 * m5. As Figure 5 shown, each arrow in the decoding search graph of the background model has its corresponding weight value. Among them, this weight value represents the probability that the background model selects this element. The probability corresponding to the second sequence is the probability L of the path "you" - "good" - "small" - "card" formed by the 4 elements "you", "good", "small", and "card" in the background model bg = n1 * n2 * n3 * n4 * n5. The electronic device can calculate the similarity between the first sequence and the second sequence by using L kws / L bg as the similarity between the first sequence and the second sequence.

[0099] In some other embodiments, when the information to be recognized is blood pressure data or heart rate, the electronic device can determine the similarity between the first sequence and the second sequence. Exemplarily, the electronic device can compare the elements in the first sequence with the elements in the second sequence to determine the similarity between the first sequence and the second sequence.

[0100] S305: When the similarity meets the set threshold condition, the electronic device outputs the first sequence as the keyword recognition result.

[0101] In some embodiments, after obtaining the similarity, the electronic device can control the output of the keyword recognition result by determining whether the similarity meets the set threshold condition. Among them, the set threshold condition can be set manually according to actual needs. The threshold in the set threshold condition can be related to the false alarm rate, that is, the user can set the relevant threshold condition according to the actual need for the false alarm rate. For example, when the information to be recognized contains keywords, the similarity between the first sequence and the second sequence is infinitely close to 1. Considering the loss of calculation accuracy and other situations, usually in actual situations, the similarity between the first sequence and the second sequence will not be equal to 1. Another example is that when the information to be recognized does not contain keywords, the similarity between the first sequence and the second sequence must be less than 1. The user can set the set threshold condition according to this situation to maintain the balance between the recognition accuracy rate and the false alarm rate.

[0102] In the embodiments of the present application, the user can directly set the threshold condition according to the actual usage requirements. Exemplarily, the user can set the threshold condition in the configuration user interface (UI) corresponding to the set threshold condition. In addition, the user can also configure the threshold condition through the application program interface (API) of the threshold condition. Among them, the false alarm rate and the threshold value in the threshold condition have a linear negative correlation. For example, when the false alarm rate is less than 0.5%, the user can configure the threshold condition as the similarity is greater than 0.995.

[0103] As an example, when the false alarm rate is X, the set threshold conditions include but are not limited to the following situations:

[0104] Situation 1: The similarity is greater than or equal to 1 - X;

[0105] Situation 2: The similarity is less than or equal to 1 and greater than or equal to 1 - X;

[0106] Situation 3: When X is between M and N, the similarity is greater than or equal to 1 - N and less than or equal to 1 - M. Here, N is greater than M.

[0107] When the electronic device determines that the similarity meets the set threshold conditions, it determines the information to be recognized as a keyword and outputs the first sequence as the keyword recognition result. As another example, when the similarity does not meet the set threshold conditions, it is determined that the information to be recognized does not contain a keyword, and then the keyword recognition result is determined to be empty. In this way, the electronic device will only output the first sequence as the keyword recognition result when the first sequence and the second sequence meet the set threshold conditions, so that the electronic device can perform related functions.

[0108] As an example, after the user configures the set threshold conditions, the embodiment of the present application can determine the keyword recognition result through a binary decision. Among them, the decision-making judgment can be expressed by the following formula:

[0109]

[0110] Among them, R likelihood is the similarity, L kws represents the probability corresponding to the first sequence output by the keyword model, and L bg represents the probability corresponding to the second sequence output by the background model. When the decision-making judgment outputs 1, it means that the first sequence is the same as the second sequence, that is, the information to be recognized includes a keyword. At this time, the electronic device will output the first sequence as the keyword recognition result. When the decision-making judgment outputs 0, it means that the information to be recognized does not include a keyword. At this time, the electronic device will not output the keyword recognition result, or the output keyword recognition result is empty. In this way, it is possible to ensure the balance between the keyword recognition accuracy rate and the false alarm rate in different scenarios and different application environments.

[0111] Based on the content shown in the above embodiments, the electronic device determines whether to output the keyword recognition result by determining whether the similarity between the first sequence and the second sequence meets the set threshold conditions. In this way, in the changes of different scenarios, different application environments, different user requirements, etc., it is convenient for the user to directly and simply adjust the set threshold conditions to limit the false alarm rate of keyword recognition, so that the false alarm rate will not change with the environment, thus ensuring the balance between the keyword recognition accuracy rate and the false alarm rate, and further ensuring the stable availability of the keyword recognition system.

[0112] The content executed by the electronic device shown in the above embodiments can be executed by an application in the electronic device. Figure 6 The figure is an application architecture diagram of an electronic device provided by an embodiment of the present application. The electronic device includes a feature extraction module and a decoding graph module.

[0113] Among them, the feature extraction module is used to extract features from the information to be recognized to obtain the feature information of the information to be recognized.

[0114] The decoding graph module is used to perform keyword recognition on the feature information to obtain a keyword recognition result.

[0115] As Figure 7 shown, the decoding graph module may include a keyword model module, a background model module, a likelihood ratio module, and a decision module.

[0116] Among them, the keyword model module is used to construct a keyword model, and perform keyword recognition on the feature information of the information to be recognized through the keyword model, and output a sequence corresponding to the maximum likelihood probability. Among them, the sequence corresponding to the maximum likelihood probability is the sequence with the highest score in the keyword model.

[0117] The background model module is used to construct a background model, and perform recognition on the feature information of the information to be recognized through the background model, and output a sequence corresponding to the maximum likelihood probability.

[0118] The likelihood ratio module is used to determine the similarity between the sequence output by the keyword model module and the sequence output by the background model module.

[0119] The decision module is used to make a decision on the output of the keyword recognition result according to the similarity.

[0120] In some embodiments, the decoding graph module of the embodiment of the present application can be trained in the following manner.

[0121] As Figure 8 shown, the embodiment of the present application can input sample features into the decoding graph module to train the decoding graph module. Among them, the sample features include sample keyword features and sample non-keyword features.

[0122] The electronic device can construct a keyword model according to the first modeling unit corresponding to the sample keyword. And the electronic device can also construct a background model of the full search space or the non-keyword space based on the second modeling unit. Among them, the construction process of the keyword model is the same as the process of constructing the keyword model in step S302 above, and the construction process of the background model is the same as the process of constructing the background model in step S303 above, and will not be described in detail here.

[0123] After constructing the keyword model and the background model, the electronic device places them into the keyword model module and the background model module in the decoding graph module respectively. After the electronic device inputs the sample features into the keyword model module, the keyword module performs keyword recognition on the sample features to obtain at least one sample sequence, and outputs the sample sequence with the maximum likelihood probability as the first sequence. Similarly, after the electronic device inputs the sample features into the background model module, the background model performs recognition on the sample features to obtain at least one sample sequence, and outputs the sample sequence with the maximum likelihood probability as the second sequence.

[0124] The keyword model module outputs the first sequence to the likelihood ratio module, and the background model module outputs the second sequence to the likelihood ratio module. The likelihood ratio module can calculate the ratio of the probability corresponding to the first sequence and the probability corresponding to the second sequence to obtain the likelihood probability ratio, and output the likelihood probability ratio as the similarity between the first sequence and the second sequence to the decision module. Among them, the probability corresponding to the first sequence is the probability that the elements in the keyword model form the first sequence, and the probability corresponding to the second sequence is the probability that the elements in the background model form the second sequence. For example, the probability corresponding to the first sequence can be the likelihood probability of the first sequence, and the probability corresponding to the second sequence can be the likelihood probability of the second sequence.

[0125] The decision module can decide the output of the keyword recognition result by judging whether the similarity meets the set threshold condition. Among them, the user can customize the threshold condition in the decision module through the UI interface or the corresponding API. The electronic device can adjust the decoding graph module according to the keyword recognition result, and then complete the training process of the decoding graph module.

[0126] Based on the content shown in the above embodiments, the user can customize the set threshold condition of the decision module in the decoding graph module according to the actual situation, so that the electronic device can output the keyword recognition result according to the user's needs, and then can achieve the balance between the accuracy rate and the false alarm rate of the keyword recognition system in any environment, meet the performance stability of the user's actual needs, and have user configurability.

[0127] Figure 9 This is a complete flow schematic diagram of a keyword recognition method provided by this application. As Figure 9 shown, the method includes the following steps:

[0128] S901: The electronic device obtains the information to be recognized.

[0129] Among them, the information to be recognized can be voice information.

[0130] S902: The electronic device extracts features from the information to be recognized to obtain the feature information of the information to be recognized.

[0131] S903: The electronic device inputs the feature information into the keyword model, performs keyword recognition on the feature information based on the keyword model, and obtains at least one first candidate sequence.

[0132] Among them, the keyword model is constructed based on the first modeling unit, and the first modeling unit includes at least one of the pronunciation, phoneme, and phrase corresponding to the keyword set.

[0133] S904: The electronic device takes the first candidate sequence corresponding to the maximum likelihood probability as the first sequence.

[0134] S905: The electronic device inputs the feature information into the background model, performs recognition on the feature information based on the background model, and obtains at least one second candidate sequence.

[0135] Among them, the background model is constructed based on the second modeling unit, and the second modeling unit includes at least one of the pronunciation, phoneme, and phrase corresponding to the full set or non-keyword set. Among them, the number of pronunciation elements, phoneme elements, or phrase elements in the background model is greater than the number of pronunciation elements, phoneme elements, or phrase elements in the keyword model.

[0136] S906: The electronic device takes the second candidate sequence corresponding to the maximum likelihood probability as the second sequence.

[0137] S907: The electronic device takes the ratio of the probability corresponding to the first sequence to the probability corresponding to the second sequence as the similarity between the first sequence and the second sequence. Among them, the probability corresponding to the first sequence represents the probability that the elements in the keyword model form the first sequence, and the probability corresponding to the second sequence represents the probability that the elements in the background model form the second sequence. The closer the ratio of the probability corresponding to the first sequence to the probability corresponding to the second sequence is to 1, the higher the similarity between the first sequence and the second sequence.

[0138] S908: The electronic device determines whether the similarity meets the set threshold condition. If so, it executes step S909; if not, it executes step S910.

[0139] S909: The electronic device outputs the first sequence as the keyword recognition result.

[0140] When the electronic device outputs the first sequence as the keyword result, it determines that the information to be recognized is a keyword and executes the relevant function. For example, when the keyword is the wake-up keyword of the electronic device, when the electronic device determines that the information to be recognized is a keyword, it wakes up the electronic device.

[0141] S910: The electronic device determines that the keyword recognition result is empty.

[0142] Based on Figure 9As shown in the content, the electronic device can determine whether the similarity between the first sequence and the second sequence meets the set threshold condition, so that the electronic device can output the recognition result according to the user's needs, and the user can also freely modify the set threshold condition according to his own needs, ensuring the balance between the keyword recognition accuracy and the false alarm rate in different scenarios and different application environments, thereby ensuring the stable availability of the keyword recognition system.

[0143] Based on the above content and the same concept, Figure 10 FIG. 5 is a schematic structural diagram of a possible keyword recognition device 1000 provided by the present application, and the keyword recognition device 1000 is applied to an electronic device.

[0144] In some embodiments, the electronic device includes the keyword recognition device 1000, or the electronic device is the keyword recognition device 1000. The keyword recognition device 1000 can be used to implement the functions of the electronic device in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments.

[0145] The recognition module 1001 is configured to obtain the information to be recognized; decode the information to be recognized through a keyword set to obtain a first sequence; decode the information to be recognized through a full set or a non-keyword set to obtain a second sequence;

[0146] The determination module 1002 is configured to determine the similarity between the first sequence and the second sequence; when the similarity meets the set threshold condition, output the first sequence as the keyword recognition result.

[0147] Optionally, when the similarity does not meet the set threshold condition, the determination module 1002 may not output the keyword recognition result.

[0148] An optional implementation manner is that the recognition module 1001 is specifically configured to:

[0149] Extract features from the information to be recognized to obtain the feature information of the information to be recognized;

[0150] Input the feature information into a keyword model, and perform keyword recognition on the feature information based on the keyword model to obtain at least one first candidate sequence;

[0151] Use the first candidate sequence corresponding to the maximum likelihood probability as the first sequence.

[0152] An optional implementation manner is that the keyword model is constructed based on a first modeling unit, and the first modeling unit includes at least one of the pronunciation, phoneme, and phrase corresponding to the keyword set.

[0153] An alternative implementation is that the recognition module 1001 is specifically configured to:

[0154] Extract features from the information to be recognized to obtain the feature information of the information to be recognized;

[0155] Input the feature information into the background model, and recognize the feature information based on the background model to obtain at least one second candidate sequence;

[0156] Take the second candidate sequence corresponding to the maximum likelihood probability as the second sequence.

[0157] An alternative implementation is that the background model is constructed based on a second modeling unit, and the second modeling unit includes at least one of phonetic sounds, phonemes, and phrases.

[0158] An alternative implementation is that the determination module 1002 is specifically configured to:

[0159] Determine the similarity according to the probability corresponding to the first sequence and the probability corresponding to the second sequence; the probability corresponding to the first sequence represents the probability of training the first sequence through the elements in the keyword set, and the probability corresponding to the second sequence represents the probability of training the second sequence through the elements in the full set or the non-keyword set.

[0160] Based on the above content and the same concept, the present application provides an electronic device, including a processor and a memory. The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the electronic device executes the steps executed by the electronic device in the above method embodiment.

[0161] Based on the above content and the same concept, the present application provides a computer-readable storage medium, in which a computer program or instruction is stored. When the computer program or instruction is executed by a computing device, the computing device executes the steps executed by the electronic device in the above method embodiment.

[0162] Based on the above content and the same concept, the present application provides a computer program product, which includes a computer program or instruction. When the computer program or instruction is executed by a computing device, the computing device executes the steps executed by the electronic device in the above method embodiment.

[0163] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0164] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0165] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0166] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0167] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A keyword recognition method, characterized in that, the method includes: Obtaining the information to be recognized; Decoding the information to be recognized through a keyword set to obtain a first sequence; Decoding the information to be recognized through a full set or a non-keyword set to obtain a second sequence; Determining the similarity between the first sequence and the second sequence; when the similarity meets the set threshold condition, outputting the first sequence as the keyword recognition result.

2. The method according to claim 1, characterized in that, the decoding the information to be recognized through a keyword set to obtain a first sequence includes: Performing feature extraction on the information to be recognized to obtain the feature information of the information to be recognized; Inputting the feature information into a keyword model, and performing keyword recognition on the feature information based on the keyword model to obtain at least one first candidate sequence; the keyword model is constructed based on the keyword set; Taking the first candidate sequence corresponding to the maximum likelihood probability as the first sequence.

3. The method according to claim 2, characterized in that, the keyword model is constructed based on a first modeling unit, and the first modeling unit includes at least one of the pronunciation, phoneme, and phrase corresponding to the keyword set.

4. The method according to any one of claims 1-3, characterized in that, the decoding the information to be recognized through a full set or a non-keyword set to obtain a second sequence includes: Performing feature extraction on the information to be recognized to obtain the feature information of the information to be recognized; Inputting the feature information into a background model, and performing recognition on the feature information based on the background model to obtain at least one second candidate sequence; the background model is constructed based on the full set or the non-keyword set; Taking the second candidate sequence corresponding to the maximum likelihood probability as the second sequence.

5. The method according to claim 4, characterized in that, the background model is constructed based on a second modeling unit, and the second modeling unit includes at least one of the pronunciation, phoneme, and phrase.

6. The method according to any one of claims 1-5, characterized in that, the determining the similarity between the first sequence and the second sequence includes: Determining the similarity according to the probability corresponding to the first sequence and the probability corresponding to the second sequence; the probability corresponding to the first sequence represents the probability of training the first sequence through the elements in the keyword set, and the probability corresponding to the second sequence represents the probability of training the second sequence through the elements in the full set or the non-keyword set.

7. A keyword recognition device, characterized in that, the device includes: An identification module, configured to obtain the information to be recognized; decode the information to be recognized through a keyword set to obtain a first sequence; decode the information to be recognized through a full set or a non-keyword set to obtain a second sequence; A determination module, configured to determine the similarity between the first sequence and the second sequence; when the similarity meets the set threshold condition, outputting the first sequence as the keyword recognition result.

8. The device according to claim 7, characterized in that, The recognition module is specifically configured to: Extract features from the information to be recognized to obtain the feature information of the information to be recognized; Input the feature information into a keyword model, and perform keyword recognition on the feature information based on the keyword model to obtain at least one first candidate sequence; Use the first candidate sequence corresponding to the maximum likelihood probability as the first sequence.

9. The apparatus according to claim 8, wherein, the keyword model is constructed based on a first modeling unit, and the first modeling unit includes at least one of the pronunciation, phoneme, and phrase corresponding to the keyword set.

10. The apparatus according to any one of claims 7-9, wherein, the recognition module is specifically configured to: Extract features from the information to be recognized to obtain the feature information of the information to be recognized; Input the feature information into a background model, and perform recognition on the feature information based on the background model to obtain at least one second candidate sequence; Use the second candidate sequence corresponding to the maximum likelihood probability as the second sequence.

11. The apparatus according to claim 10, wherein, the background model is constructed based on a second modeling unit, and the second modeling unit includes at least one of pronunciation, phoneme, and phrase.

12. The apparatus according to any one of claims 7-11, wherein, the determination module is specifically configured to: Determine the similarity according to the probability corresponding to the first sequence and the probability corresponding to the second sequence; the probability corresponding to the first sequence represents the probability of training the first sequence through the elements in the keyword set, and the probability corresponding to the second sequence represents the probability of training the second sequence through the elements in the full set or the non-keyword set.

13. An electronic device, wherein, comprising: One or more processors; One or more memories; Wherein, the one or more memories store one or more computer programs, and the one or more computer programs include instructions, and when the instructions are executed by the one or more processors, the electronic device executes the method according to any one of claims 1 to 6.

14. A computer-readable storage medium, wherein, the computer-readable storage medium is used to store a computer program, and when the computer program runs on a computer, the computer executes the method according to any one of claims 1 to 6.

15. A computer program product, wherein, comprising a computer program, and when the computer program runs on a computer, the computer executes the method according to any one of claims 1 to 6 above.

16. A chip, wherein, the chip is coupled to a memory, and the chip reads the computer program stored in the memory and executes the method according to any one of claims 1 to 6.

Images