Wake-up method, device, electronic device and storage medium

Abstract

The present application relates to a wake-up method, device, electronic device and storage medium, which are applied to the field of speech recognition technology. The wake-up method includes: obtaining the speech frame corresponding to the wake-up speech in real time; when it is determined that the smart device cannot be woken up based on the speech frames in the first time period, the speech frames in the first time period are determined as the first target sequence; after a preset time interval at the end of the first time period, when it is determined that the smart device cannot be woken up based on the speech frames in the second time period, the speech frames in the second time period are determined as the second target sequence; according to the first target sequence and the second target sequence, it is determined whether to wake up the smart device. The present application can improve the wake-up rate of secondary wake-up.

Description

Technical Field

[0001] The present application relates to the field of speech recognition technology, and in particular to a wake-up method, device, electronic device and storage medium. Background Art

[0002] With the development of smart devices, voice interaction has achieved unprecedented growth. Wake-up is the most important part of it and is the portal for users to interact with smart devices. In the presence of background noise, if the user fails to wake up the smart device when shouting the wake-up word for the first time, the user can shout the wake-up word again to wake up the smart device for the second time. However, in the existing wake-up algorithms, the wake-up rate of the second wake-up is low. Summary of the invention

[0003] In order to solve the above technical problem or at least partially solve the above technical problem, the present application provides a wake-up method, device, electronic device and storage medium.

[0004] According to a first aspect of the present application, a wake-up method is provided, comprising:

[0005] Acquire the voice frame corresponding to the wake-up voice in real time;

[0006] When it is determined that the smart device cannot be woken up based on the voice frames in the first time period, determining the voice frames in the first time period as a first target sequence;

[0007] After a preset time interval from the end of the first time period, when it is determined based on the voice frames in the second time period that the smart device cannot be woken up, the voice frames in the second time period are determined as a second target sequence;

[0008] Determine whether to wake up the smart device according to the first target sequence and the second target sequence.

[0009] Optionally, determining whether to wake up the smart device according to the first target sequence and the second target sequence includes:

[0010] Cross-correlating the first target sequence with the second target sequence to obtain a cross-correlation value;

[0011] If the mutual correlation value is greater than or equal to a preset mutual correlation threshold, waking up the smart device;

[0012] If the mutual correlation value is less than the preset mutual correlation threshold, the smart device is not woken up.

[0013] Optionally, the method further includes:

[0014] Acquire first fundamental frequency information of the first target sequence and second fundamental frequency information of the second target sequence;

[0015] If the similarity between the first baseband information and the second baseband information is greater than or equal to a similarity threshold, performing the step of cross-correlating the first target sequence and the second target sequence;

[0016] If the similarity between the first baseband information and the second baseband information is less than the similarity threshold, the smart device is not woken up.

[0017] Optionally, judging whether to wake up the smart device based on the voice frames in the first time period or the second time period includes:

[0018] Inputting the speech frames in the first time period or the second time period into a pre-trained wake-up model to obtain a wake-up score, where the wake-up score is used to indicate the probability that the text information corresponding to the speech frames in the first time period or the second time period belongs to the wake-up word;

[0019] If the wake-up score is greater than or equal to a preset score threshold, determine to wake up the smart device;

[0020] If the wake-up score is less than the preset score threshold, determine not to wake up the smart device.

[0021] Optionally, the wake-up method further includes:

[0022] Selecting a first speech frame sequence and a second speech frame sequence having a preset length from the first target sequence and the second target sequence respectively;

[0023] The step of determining whether to wake up the smart device according to the first target sequence and the second target sequence includes:

[0024] Determine whether to wake up the smart device according to the first voice frame sequence and the second voice frame sequence.

[0025] Optionally, the wake-up method further includes:

[0026] Acquire a plurality of first sample wake-up voices after one wake-up;

[0027] The value of the preset length is determined according to the length of the speech frame corresponding to the first sample wake-up speech.

[0028] Optionally, the wake-up method further includes:

[0029] Acquire multiple second sample wake-up voices that have been woken up at least twice;

[0030] The preset time interval is determined according to the time interval between the end frame of the first wake-up and the start frame of the second wake-up in the second sample wake-up voice.

[0031] According to a second aspect of the present application, a wake-up device is provided, including:

[0032] A voice frame real-time acquisition module is used to acquire the voice frame corresponding to the wake-up voice in real time;

[0033] A first target sequence determination module, configured to, when it is determined based on the voice frames in the first time period that the intelligent device cannot be woken up, determine the voice frames in the first time period as the first target sequence;

[0034] A second target sequence determination module is configured to, after a preset time interval from the end of the first time period, determine the voice frames in the second time period as a second target sequence when it is determined that the smart device cannot be woken up based on the voice frames in the second time period;

[0035] The wake-up judgment module is used to determine whether to wake up the smart device according to the first target sequence and the second target sequence.

[0036] Optionally, the wake-up judgment module is specifically used to cross-correlate the first target sequence and the second target sequence to obtain a cross-correlation value; if the cross-correlation value is greater than or equal to a preset cross-correlation threshold, the smart device is woken up; if the cross-correlation value is less than the preset cross-correlation threshold, the smart device is not woken up.

[0037] Optionally, the awakening device further includes:

[0038] A fundamental frequency information acquisition module, used to acquire first fundamental frequency information of the first target sequence and second fundamental frequency information of the second target sequence;

[0039] The wake-up judgment module is specifically configured to execute the step of cross-correlating the first target sequence with the second target sequence if the similarity between the first baseband information and the second baseband information is greater than or equal to a similarity threshold;

[0040] The wake-up judgment module is specifically configured to not wake up the smart device if the similarity between the first baseband information and the second baseband information is less than the similarity threshold.

[0041] Optionally, the first target sequence determination module determines whether to wake up the smart device based on the voice frames in the first time period through the following steps:

[0042] Inputting the speech frames in the first time period into a pre-trained wake-up model to obtain a wake-up score, where the wake-up score is used to indicate the probability that the text information corresponding to the speech frames in the first time period belongs to the wake-up word;

[0043] If the wake-up score is greater than or equal to a preset score threshold, determine to wake up the smart device;

[0044] If the wake-up score is less than the preset score threshold, determine not to wake up the smart device;

[0045] The second target sequence determination module determines whether to wake up the smart device based on the voice frames in the second time period through the following steps:

[0046] Inputting the speech frames in the second time period into a pre-trained wake-up model to obtain a wake-up score, where the wake-up score is used to indicate the probability that the text information corresponding to the speech frames in the second time period belongs to the wake-up word;

[0047] If the wake-up score is greater than or equal to a preset score threshold, determine to wake up the smart device;

[0048] If the wake-up score is less than the preset score threshold, determine not to wake up the smart device.

[0049] Optionally, the awakening device further includes:

[0050] A speech frame sequence selection module, used to select a first speech frame sequence and a second speech frame sequence with a preset length from the first target sequence and the second target sequence respectively;

[0051] The wake-up judgment module is specifically used to determine whether to wake up the smart device according to the first voice frame sequence and the second voice frame sequence.

[0052] Optionally, the awakening device further includes:

[0053] A first sample wake-up voice acquisition module, used to acquire a plurality of first sample wake-up voices after one wake-up;

[0054] The voice frame length determination module is used to determine the value of the preset length according to the length of the voice frame corresponding to the first sample wake-up voice.

[0055] Optionally, the awakening device further includes:

[0056] A second sample wake-up voice acquisition module, used to acquire a plurality of second sample wake-up voices that have been woken up at least twice;

[0057] The preset time interval determination module is used to determine the preset time interval according to the time interval between the end frame of the first wake-up and the start frame of the second wake-up in the second sample wake-up voice.

[0058] According to a third aspect of the present application, an electronic device is provided, comprising: a processor, wherein the processor is configured to execute a computer program stored in a memory, wherein the computer program implements the method described in the first aspect when executed by the processor.

[0059] According to a fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method described in the first aspect is implemented.

[0060] According to a fifth aspect of the present application, a computer program product is provided. When the computer program product is run on a computer, the computer is enabled to execute the method described in the first aspect.

[0061] Compared with the prior art, the technical solution provided by the embodiments of the present application has the following advantages:

[0062] When the user fails to wake up the smart device for the first time, a second wake-up is usually performed. Since the two wake-ups usually have a certain time interval, the first target sequence and the second target sequence with a preset time interval can be extracted. That is, it can be pre-determined that the user has performed a second wake-up. Furthermore, during the second wake-up, the same wake-up algorithm as the first wake-up is no longer used, that is, the first target sequence is treated as noise, and the second target sequence is processed by the wake-up algorithm. Instead, the correlation between the two wake-ups is considered, and whether to wake up the smart device is determined based on the first target sequence and the second target sequence, thereby improving the wake-up rate of the second wake-up and enhancing the user's interactive experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0064] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0065] Figure 1 A schematic diagram of an application scenario of the wake-up method in an embodiment of the present application;

[0066] Figure 2 A flowchart of the wake-up method in an embodiment of the present application;

[0067] Figure 3 This is another flow chart of the wake-up method in the embodiment of the present application;

[0068] Figure 4 A schematic diagram of a structure of a wake-up device in an embodiment of the present application;

[0069] Figure 5 A schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0070] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the scheme of the present application will be further described below. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0071] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only part of the embodiments of the present application, rather than all of the embodiments.

[0072] Figure 1 Schematic diagram of an application scenario of the wake-up method in the embodiment of the present application. Figure 1 As shown, the smart speaker has a voice wake-up function, and the user can shout a specific wake-up word to wake up the smart speaker. This application can improve the wake-up rate of the second wake-up when the user fails to wake up the speaker the first time.

[0073] The following first introduces the wake-up method of the embodiment of the present application in detail.

[0074] See also Figure 2 , Figure 2 A flowchart of the wake-up method in an embodiment of the present application may include the following steps:

[0075] Step S210, obtaining a voice frame corresponding to the wake-up voice in real time.

[0076] Step S220: When it is determined that the smart device cannot be woken up based on the voice frames in the first time period, the voice frames in the first time period are determined as a first target sequence.

[0077] Step S230, after a preset time interval from the end of the first time period, when it is determined that the smart device cannot be woken up based on the voice frames in the second time period, the voice frames in the second time period are determined as the second target sequence.

[0078] Step S240: Determine whether to wake up the smart device according to the first target sequence and the second target sequence.

[0079] The wake-up method of the embodiment of the present application usually performs a second wake-up when the user fails to wake up the smart device for the first time. Since the two wake-ups usually have a certain time interval, the first target sequence and the second target sequence with a preset time interval can be extracted. That is, it can be pre-determined that the user has performed a second wake-up. Furthermore, during the second wake-up, the same wake-up algorithm as the first wake-up is no longer used, that is, the first target sequence is treated as noise, and the second target sequence is processed by the wake-up algorithm. Instead, the correlation between the two wake-ups is considered, and whether to wake up the smart device is determined based on the first target sequence and the second target sequence, thereby improving the wake-up rate of the second wake-up and enhancing the user's interactive experience.

[0080] See also Figure 3 , Figure 3 This is another flowchart of the wake-up method in the embodiment of the present application, which may include the following steps:

[0081] Step S302, obtaining the speech frame corresponding to the wake-up speech in real time.

[0082] When a user performs voice wake-up on a smart device (such as a smart TV, smart speaker, etc.) using a wake-up word, a corresponding wake-up voice may be generated. The wake-up voice usually includes multiple voice frames, and the smart device may obtain the multiple voice frames in real time.

[0083] Step S304: when it is determined that the smart device cannot be woken up based on the voice frames in the first time period, the voice frames in the first time period are determined as a first target sequence.

[0084] It should be noted that since the process of the user shouting the wake-up word usually corresponds to a time period (for example, 1s), multiple voice frames can be included in the time period. Therefore, when the time the user shouts the wake-up word reaches the preset time period, that is, the length of the voice frame in the preset time period reaches the preset length, the smart device can be woken up. In other words, if the time the user shouts the wake-up word does not reach the preset time period, or the length of the voice frame does not reach the preset length, the smart device cannot be woken up. Among them, the preset time period and the preset length can be determined according to the length of the wake-up word. The longer the length of the wake-up word, the larger the preset time period and the preset length will be.

[0085] In an embodiment of the present application, since the voice frames are acquired in real time, that is, the smart device acquires the first voice frame, the second voice frame..., until the last voice frame in sequence. The present application can detect the voice frame boundary when the user shouts the wake-up word based on VAD (voice activity detection) technology. Based on the voice frame boundary and the preset time period, the voice frame in the first time period can be determined. Among them, the first time period can be greater than or equal to the preset time period. Determine whether the smart device can be woken up based on the voice frames in the first time period. If the smart device can be woken up, the smart device can be woken up directly. If the smart device cannot be woken up, the voice frames in the first time period are determined as the first target sequence, and the first target sequence can also be stored in the first buffer area. The first target sequence represents the voice frame sequence corresponding to the user when the smart device cannot be woken up for the first time.

[0086] Optionally, the following method can be used to determine whether to wake up the smart device based on the voice frames in the first time period: input the voice frames in the first time period into a pre-trained wake-up model to obtain a wake-up score, which is used to indicate the probability that the text information corresponding to the voice frames in the first time period belongs to the wake-up word; if the wake-up score is greater than or equal to a preset score threshold, determine to wake up the smart device; if the wake-up score is less than the preset score threshold, determine not to wake up the smart device. The preset score threshold can be a preset score threshold, such as 80%, 90%, etc.; the preset score threshold is usually set to a higher value to avoid false wake-ups.

[0087] It should be noted that, when the wake-up score is less than the preset score threshold, it can also be determined whether the wake-up score is greater than the minimum score threshold. If it is greater than the minimum score threshold, it means that the user may intend to wake up the smart device, but fails to wake up the smart device due to environmental noise and other reasons; if it is less than the minimum score threshold, it means that the user may not intend to wake up the smart device. Therefore, in some embodiments, when it is determined that the smart device cannot be woken up based on the voice frames in the first time period, and the wake-up score corresponding to the voice frames in the first time period is greater than the minimum score threshold, the voice frames in the first time period are determined as the first target sequence. In this way, the accuracy of determining the first target sequence can be improved.

[0088] Step S306, after a preset time interval from the end of the first time period, when it is determined that the smart device cannot be woken up based on the voice frames in the second time period, the voice frames in the second time period are determined as the second target sequence.

[0089] In the case where the user fails to wake up the smart device during the first wake-up, the failure to wake up the smart device for the first time may be because the wake-up was affected by other problems such as noise at that time. At this time, the wake-up score is less than the preset score threshold, and the user usually wakes up for the second time. Therefore, after the voice frame in the first time period, new voice frames can continue to be obtained. If a new voice frame can be obtained, it means that the user has performed a second wake-up. If no new voice frame is obtained, it means that the user did not perform a second wake-up, and the process ends. Of course, if the smart device obtains a new voice frame after a long time, it is considered that this wake-up is unrelated to the previous wake-up and does not belong to a second wake-up, but a new wake-up.

[0090] If the smart device can be woken up after acquiring a new voice frame, the smart device can be woken up directly. If the smart device still cannot be woken up, a second target sequence can be further acquired, that is, the voice frame corresponding to the user's second wake-up. Similar to the first target sequence, the second target sequence can also be acquired based on the VAD technology.

[0091] It should be noted that when the user performs a second wake-up, there is usually a short time interval between the two wake-ups. Therefore, the second target sequence can be obtained after a preset time interval. If the time interval between the voice frame sequence obtained in the second time period and the first target sequence is greater than or equal to the preset time interval, the obtained voice frame sequence is considered to be the second target frame sequence, indicating that the voice frame sequence corresponding to the user cannot wake up the smart device in the second wake-up. If the time interval between the voice frame sequence obtained in the second time period and the first target sequence is less than the preset time interval, it means that the obtained voice frame sequence is not the second target frame sequence, and the first target sequence can be updated by the obtained voice frame sequence, that is, the first target sequence is updated to the voice frame sequence, and the second target sequence is re-determined according to the above method. After determining the second target sequence, the second target frame sequence can be stored in the second buffer area.

[0092] Similarly, whether to wake up the smart device can be determined based on the speech frames in the second time period in the following manner: input the speech frames in the second time period into the pre-trained wake-up model to obtain a wake-up score, and the wake-up score is used to indicate the probability that the text information corresponding to the speech frames in the second time period belongs to the wake-up word; if the wake-up score is greater than or equal to the preset score threshold, determine to wake up the smart device; if the wake-up score is less than the preset score threshold, determine not to wake up the smart device. Among them, the second time period may also be greater than or equal to the preset time period. Similarly, when the wake-up score corresponding to the speech frames in the second time period is greater than the minimum score threshold, the speech frames in the second time period can be determined as the second target sequence.

[0093] Optionally, multiple second sample wake-up voices that have been awakened at least twice can be obtained; the preset time interval is determined according to the time interval between the end frame of the first awakening and the start frame of the second awakening in the second sample wake-up voice. For example, the multiple obtained time intervals are sorted from low to high, the top 10% of the time intervals are selected and averaged to obtain the preset time interval.

[0094] Among them, the second sample wake-up voice is the voice data of the second wake-up, that is, the user did not wake up the smart device when shouting the wake-up word for the first time, and the user woke it up for the second time in a short period of time. If the user did not wake up the smart device when shouting the wake-up word for the first time, and shouted the wake-up word again after a long time, then this wake-up is a new wake-up, that is, the two wake-up voices do not belong to the second sample wake-up voice. If the user says a word close to the wake-up word twice in a short period of time, the user may not intend to wake up the device at this time. If the wake-up score is in the preset range (that is, greater than the minimum score threshold and less than the preset score threshold), it can also be determined as a second wake-up.

[0095] Step S308: acquiring first fundamental frequency information of the first target sequence and second fundamental frequency information of the second target sequence, and determining whether the similarity between the first fundamental frequency information and the second fundamental frequency information is greater than or equal to a similarity threshold.

[0096] Fundamental frequency refers to the basic frequency of a user when pronouncing a word, and the fundamental frequencies corresponding to different users are usually different. After obtaining the first target sequence and the second target sequence, the first fundamental frequency information and the second fundamental frequency information corresponding to the first target sequence and the second target sequence can be obtained respectively. If the similarity between the first fundamental frequency information and the second fundamental frequency information is greater than or equal to the similarity threshold, it means that the first target sequence and the second target sequence are voice data of the same user, and step S310 can be executed; if the similarity between the first fundamental frequency information and the second fundamental frequency information is less than the similarity threshold, it means that the first target sequence and the second target sequence are not voice data of the same user, and step S314 can be executed, that is, the smart device is not woken up.

[0097] Step S310: cross-correlate the first target sequence and the second target sequence to obtain a cross-correlation value, and determine whether the cross-correlation value is greater than or equal to a preset cross-correlation threshold.

[0098] The mutual correlation value indicates the degree of correlation between the first target sequence and the second target sequence. If the mutual correlation value is greater than or equal to the preset mutual correlation threshold, it indicates that the first target sequence and the second target sequence are correlated. For example, the first target sequence and the second target sequence are both voice data corresponding to the user's wake-up word, and step S312 can be executed, that is, waking up the smart device; if the mutual correlation value is less than the preset mutual correlation threshold, it indicates that the first target sequence and the second target sequence are not correlated, and step S314 can be executed, that is, not waking up the smart device.

[0099] When the user fails to wake up the smart device for the first time, if the wake-up score is within the preset range, the user may have intended to wake up the smart device, but failed to do so due to other issues such as noise. If a second voice is detected later, it can be further determined that the user realizes that the smart device cannot be woken up for the first time and wants to wake up the device with the second wake-up voice. The system will compare the second voice with the first voice. If the two are highly similar and the scores are within the preset range, it is determined that the user wants to wake up the smart device by repeating the wake-up voice, and finally wake up the smart device.

[0100] In the embodiment of the present application, the first target sequence and the second target sequence may be speech frame sequences of unequal lengths, and therefore, the first speech frame sequence and the second speech frame sequence of preset lengths may be selected from the first target sequence and the second target sequence, respectively. Optionally, a speech frame of preset length located between the first target sequence and the second target sequence may be selected, or speech frames of preset length may be selected backward from the start frames of the first target sequence and the second target sequence, or speech frames of preset length may be selected forward from the end frames of the first target sequence and the second target sequence.

[0101] Afterwards, whether to wake up the smart device is determined based on the first voice frame sequence and the second voice frame sequence. In this way, when the first voice frame sequence and the second voice frame sequence of the same length are subjected to correlation judgment, the accuracy of the correlation judgment can be improved, thereby improving the wake-up rate of the secondary wake-up.

[0102] Optionally, the value of the preset length can be determined by:

[0103] Obtain multiple first sample wake-up voices that have been awakened once, and determine the value of the preset length according to the length of the voice frame corresponding to the first sample wake-up voice. For example, the length of the voice frame corresponding to each first sample wake-up voice can be obtained, and the length of the voice frame can be sorted from high to low, and the length of the first 10% can be selected and averaged to obtain the value of the preset length.

[0104] Step S312, waking up the smart device.

[0105] Step S314: Do not wake up the smart device.

[0106] The wake-up method of the embodiment of the present application usually performs a second wake-up when the user fails to wake up the smart device for the first time. Since the two wake-ups usually have a certain time interval, when it is determined that the smart device cannot be woken up according to the wake-up model, the first target sequence and the second target sequence with a certain time interval are extracted according to the wake-up score output by the wake-up model, that is, the speech frame sequences corresponding to the two wake-ups performed by the user, thereby pre-determining that the user has performed a second wake-up. Furthermore, based on the fundamental frequency information and the cross-correlation value of the first target sequence and the second target sequence, it is determined whether to wake up the smart device, thereby effectively improving the wake-up rate of the user's second wake-up and enhancing the user's interactive experience.

[0107] Corresponding to the above method embodiment, the present application embodiment provides a wake-up device, see Figure 4 , the awakening device 400 comprises:

[0108] A voice frame real-time acquisition module 410 is used to acquire the voice frame corresponding to the wake-up voice in real time;

[0109] A first target sequence determination module 420, configured to determine the speech frames in the first time period as a first target sequence when it is determined that the smart device cannot be woken up based on the speech frames in the first time period;

[0110] A second target sequence determination module 430 is configured to determine the voice frames in the second time period as a second target sequence after a preset time interval from the end of the first time period when it is determined based on the voice frames in the second time period that the smart device cannot be woken up;

[0111] The wake-up determination module 440 is used to determine whether to wake up the smart device according to the first target sequence and the second target sequence.

[0112] Optionally, the wake-up judgment module 440 is specifically used to cross-correlate the first target sequence and the second target sequence to obtain a cross-correlation value; if the cross-correlation value is greater than or equal to a preset cross-correlation threshold, the smart device is woken up; if the cross-correlation value is less than the preset cross-correlation threshold, the smart device is not woken up.

[0113] Optionally, the awakening device 400 further includes:

[0114] A baseband information acquisition module, used to acquire first baseband information of a first target sequence and second baseband information of a second target sequence;

[0115] The wake-up judgment module 440 is specifically used to execute the step of cross-correlating the first target sequence with the second target sequence if the similarity between the first baseband information and the second baseband information is greater than or equal to the similarity threshold; if the similarity between the first baseband information and the second baseband information is less than the similarity threshold, not wake up the smart device.

[0116] Optionally, the first target sequence determination module 420 determines whether to wake up the smart device based on the voice frames in the first time period through the following steps:

[0117] Inputting the speech frames in the first time period into the pre-trained wake-up model to obtain a wake-up score, where the wake-up score is used to indicate the probability that the text information corresponding to the speech frames in the first time period belongs to the wake-up word;

[0118] If the wake-up score is greater than or equal to a preset score threshold, determine to wake up the smart device;

[0119] If the wake-up score is less than a preset score threshold, determine not to wake up the smart device;

[0120] The second target sequence determination module 430 determines whether to wake up the smart device based on the voice frames in the second time period through the following steps:

[0121] Inputting the speech frames in the second time period into the pre-trained wake-up model to obtain a wake-up score, where the wake-up score is used to indicate the probability that the text information corresponding to the speech frames in the second time period belongs to the wake-up word;

[0122] If the wake-up score is greater than or equal to a preset score threshold, determine to wake up the smart device;

[0123] If the wake-up score is less than the preset score threshold, it is determined not to wake up the smart device.

[0124] Optionally, the awakening device 400 further includes:

[0125] A speech frame sequence selection module, used to select a first speech frame sequence and a second speech frame sequence with a preset length from the first target sequence and the second target sequence respectively;

[0126] The wake-up judgment module is specifically used to determine whether to wake up the smart device according to the first voice frame sequence and the second voice frame sequence.

[0127] Optionally, the awakening device 400 further includes:

[0128] A first sample wake-up voice acquisition module, used to acquire a plurality of first sample wake-up voices after one wake-up;

[0129] The voice frame length determination module is used to determine the value of the preset length according to the length of the voice frame corresponding to the first sample wake-up voice.

[0130] Optionally, the awakening device 400 further includes:

[0131] A second sample wake-up voice acquisition module, used to acquire a plurality of second sample wake-up voices that have been woken up at least twice;

[0132] The preset time interval determination module is used to determine the preset time interval according to the time interval between the end frame of the first awakening and the start frame of the second awakening in the second sample awakening voice.

[0133] The specific details of each module or unit in the above device have been described in detail in the corresponding method, so they will not be repeated here.

[0134] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0135] In an exemplary embodiment of the present application, an electronic device is further provided, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the above-mentioned wake-up method in this exemplary embodiment.

[0136] Figure 5 This is a schematic diagram of the structure of an electronic device in an embodiment of the present application. It should be noted that: Figure 5 The electronic device 500 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0137] like Figure 5 As shown, electronic device 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage part 508 into a random access memory (RAM) 503. In RAM 503, various programs and data required for system operation are also stored. Central processing unit 501, ROM 502 and RAM 503 are connected to each other via a bus 504. Input / output (I / O) interface 505 is also connected to bus 504.

[0138] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a local area network (LAN) card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed, so that a computer program read therefrom is installed into the storage section 508 as needed.

[0139] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 509, and / or installed from a removable medium 511. When the computer program is executed by the central processing unit 501, various functions defined in the device of the present application are executed.

[0140] In an embodiment of the present application, a computer-readable storage medium is further provided, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned wake-up method is implemented.

[0141] It should be noted that the computer-readable storage medium shown in the present application may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, radio frequency, etc., or any suitable combination of the above.

[0142] In an embodiment of the present application, a computer program product is also provided. When the computer program product is run on a computer, the computer executes the above-mentioned wake-up method.

[0143] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0144] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A wake-up method, characterized in that: The method comprises: Acquire the voice frame corresponding to the wake-up voice in real time; When it is determined that the smart device cannot be woken up based on the voice frames in the first time period, determining the voice frames in the first time period as a first target sequence; Within a preset time interval of the end moment of the first time period, when it is determined based on the voice frames in the second time period that the smart device cannot be woken up, the voice frames in the second time period are determined as a second target sequence; Determining whether to wake up the smart device according to the first target sequence and the second target sequence; The step of determining whether to wake up the smart device according to the first target sequence and the second target sequence includes: Cross-correlating the first target sequence with the second target sequence to obtain a cross-correlation value; If the mutual correlation value is greater than or equal to a preset mutual correlation threshold, waking up the smart device; If the mutual correlation value is less than the preset mutual correlation threshold, the smart device is not awakened; The method further includes: acquiring first fundamental frequency information of the first target sequence, and second fundamental frequency information of the second target sequence; If the similarity between the first baseband information and the second baseband information is greater than or equal to a similarity threshold, performing the step of cross-correlating the first target sequence and the second target sequence; If the similarity between the first baseband information and the second baseband information is less than the similarity threshold, the smart device is not woken up.

2. The method according to claim 1, characterized in that Determining whether to wake up the smart device based on the voice frames in the first time period or the second time period includes: Inputting the speech frames in the first time period or the second time period into a pre-trained wake-up model to obtain a wake-up score, where the wake-up score is used to indicate the probability that the text information corresponding to the speech frames in the first time period or the second time period belongs to the wake-up word; If the wake-up score is greater than or equal to a preset score threshold, determine to wake up the smart device; If the wake-up score is less than the preset score threshold, determine not to wake up the smart device.

3. The method according to claim 1, characterized in that The method further comprises: Selecting a first speech frame sequence and a second speech frame sequence having a preset length from the first target sequence and the second target sequence respectively; The step of determining whether to wake up the smart device according to the first target sequence and the second target sequence includes: Determine whether to wake up the smart device according to the first voice frame sequence and the second voice frame sequence.

4. The method according to claim 3, characterized in that The method further comprises: Acquire a plurality of first sample wake-up voices after one wake-up; The value of the preset length is determined according to the length of the speech frame corresponding to the first sample wake-up speech.

5. The method according to claim 1, characterized in that The method further comprises: Acquire multiple second sample wake-up voices that have been woken up at least twice; The preset time interval is determined according to the time interval between the end frame of the first wake-up and the start frame of the second wake-up in the second sample wake-up voice.

6. A wake-up device, characterized in that: The device comprises: A voice frame real-time acquisition module is used to acquire the voice frame corresponding to the wake-up voice in real time; A first target sequence determination module, configured to, when it is determined based on the voice frames in the first time period that the intelligent device cannot be woken up, determine the voice frames in the first time period as the first target sequence; A second target sequence determination module is configured to, within a preset time interval at the end of the first time period, determine the voice frames in the second time period as a second target sequence when it is determined that the smart device cannot be woken up based on the voice frames in the second time period; A wake-up judgment module, used to determine whether to wake up the smart device according to the first target sequence and the second target sequence; The wake-up judgment module is specifically used to cross-correlate the first target sequence and the second target sequence to obtain a cross-correlation value; if the cross-correlation value is greater than or equal to a preset cross-correlation threshold, the smart device is woken up; if the cross-correlation value is less than the preset cross-correlation threshold, the smart device is not woken up; The wake-up device further includes: a baseband information acquisition module, configured to acquire first baseband information of the first target sequence and second baseband information of the second target sequence; The wake-up judgment module is specifically configured to execute the step of cross-correlating the first target sequence with the second target sequence if the similarity between the first baseband information and the second baseband information is greater than or equal to a similarity threshold; The wake-up judgment module is specifically configured to not wake up the smart device if the similarity between the first baseband information and the second baseband information is less than the similarity threshold.

7. An electronic device, characterized in that: include: A processor, wherein the processor is used to execute a computer program stored in a memory, wherein the computer program, when executed by the processor, implements the steps of the method described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images