WO2015188459A1 - 一种终端控制方法、装置、语音控制装置及终端 - Google Patents

一种终端控制方法、装置、语音控制装置及终端 Download PDF

Info

Publication number
WO2015188459A1
WO2015188459A1 PCT/CN2014/085066 CN2014085066W WO2015188459A1 WO 2015188459 A1 WO2015188459 A1 WO 2015188459A1 CN 2014085066 W CN2014085066 W CN 2014085066W WO 2015188459 A1 WO2015188459 A1 WO 2015188459A1
Authority
WO
WIPO (PCT)
Prior art keywords
voice
segment
voice segment
terminal
module
Prior art date
Application number
PCT/CN2014/085066
Other languages
English (en)
French (fr)
Inventor
蒋志峰
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to EP14894647.8A priority Critical patent/EP3157003B1/en
Priority to US15/317,760 priority patent/US20170110131A1/en
Publication of WO2015188459A1 publication Critical patent/WO2015188459A1/zh

Links

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00—Speaker identification or verification techniques
    • G10L17/22—Interactive procedures; Man-machine interfaces
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00—Speaker identification or verification techniques
    • G10L17/02—Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/78—Detection of presence or absence of voice signals
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L2015/088—Word spotting
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G10L2015/223—Execution procedure of a spoken command

Definitions

  • the present invention relates to the field of terminal control, and in particular, to a terminal control method, a device, a voice control device, and a terminal.
  • a terminal control method a device
  • a voice control device a terminal
  • a terminal control method a device
  • a voice control device a terminal
  • a terminal control method a device
  • a voice control device a terminal
  • a terminal control method a device
  • a voice control device a voice control device
  • a terminal BACKGROUND OF THE INVENTION
  • BACKGROUND OF THE INVENTION With the enhancement of the processing capability of intelligent terminals, users have also put forward higher and higher requirements for intelligent terminals. For example, in order to liberate the hands of users, the prior art provides various voice recognition and control technologies, such as Baidu Google. Voice input, voice recognition of Keda Xunfei and siri of Apple, these applications give users better convenience, free hands without typing, and also bring great entertainment, local or based Voice intelligent recognition of Internet cloud computing
  • the embodiments of the present invention provide a terminal control method, a device, a voice control device, and a terminal, to solve the problem of complicated operation of the voice control technology in the prior art.
  • the embodiment of the present invention provides a terminal control method.
  • the terminal control method includes: after detecting that the terminal main voice system is closed, listening to a user voice including at least one voice segment, and determining whether an open attribute is present.
  • the voice segment if present, sends an open signal for enabling the primary voice system of the terminal, and stores the voice segment that is monitored after the voice segment with the open attribute, and after detecting that the terminal main voice system is successfully turned on, the voice segment is stored.
  • the voice segment is transmitted out.
  • the step of determining whether a voice segment having an open attribute occurs in the foregoing embodiment includes: sequentially acquiring content feature parameters of each voice segment in the monitored user voice, respectively matching with a preset open parameter, where the content feature parameter is used. Identifying the content of the voice segment; if the content feature parameter of a voice segment matches the preset open parameter, the voice segment has an open attribute.
  • the foregoing embodiment further includes: acquiring an identity identification parameter of each voice segment, and respectively comparing with a preset legal user identity parameter, where the identity parameter is used to identify a speaker identity of the voice segment;
  • the ON signal is sent, and the voice segment monitored after the voice segment with the ON attribute is stored.
  • the foregoing embodiment before storing the voice segment that is monitored after the voice segment having the open attribute, further includes: storing the voice segment with the open attribute.
  • the foregoing embodiment before storing the voice segment having the open attribute, further includes: determining whether the voice segment having the open attribute has a control attribute at the same time, and if yes, storing the voice segment having the open attribute.
  • the foregoing embodiment further includes: after detecting that the terminal main voice system is successfully turned on, determining whether the voice segment is being stored; if yes, continuing to store the voice segment, and after the voice segment is interrupted, terminating the storage of the subsequent voice segment If not, the storage of the subsequent voice segment is terminated; after the storage of the subsequent voice segment is terminated, the subsequent voice segment and all the stored voice segments are transmitted.
  • the terminal primary voice system sequentially processes multiple voice segments or sequentially according to the listening time of each voice segment. Multiple voice segments to the network side.
  • the embodiment of the present invention further provides a terminal control apparatus.
  • the terminal control apparatus includes: a voice monitoring module configured to listen to a user voice including at least one voice segment after detecting that the terminal main voice system is turned off.
  • the voice control device includes a monitoring module, a detection module, a control module, and a storage module, where the embodiment of the present invention is used in an actual life.
  • the monitoring module is configured to: after detecting that the terminal main voice system is turned off, listening to the user voice including the at least one voice segment, and transmitting the voice to the detecting module; and the detecting module is configured to determine whether the voice with the opening attribute is displayed in the user voice monitored by the monitoring module. Segment, and when a voice segment with an open attribute appears, triggers the control module; The control module is configured to send an open signal for starting the main voice system of the terminal when the module is triggered, and control the storage module to store the voice segment monitored by the monitoring module after the voice segment with the open attribute, and is also configured to detect the terminal.
  • the control storage module transmits the stored voice segment; the storage module is configured to store or transmit the voice segment under the control of the control module.
  • the embodiment of the present invention also provides a terminal.
  • the terminal includes a terminal main voice system and a voice control device provided by the present invention; and the voice control device is configured to: when the terminal main voice system is turned off, the monitoring includes The user voice of at least one voice segment determines whether a voice segment with an open attribute appears, and when a voice segment with an open attribute appears, the terminal main voice system is turned on, and the voice segment monitored after the voice segment having the open attribute is stored. After the terminal main voice system is turned on, the stored voice segment is handed over to the terminal main voice system.
  • the method provided by the embodiment of the present invention continuously monitors a user voice including at least one voice segment when the terminal main voice system is in a closed state, and sequentially detects each voice segment in the user voice.
  • a voice segment with an open attribute is detected, an open signal is sent to the terminal main voice system, and at the same time, the voice segment that is stored after the voice segment with the open attribute is started, and the stored voice is awake after the terminal main voice system wakes up.
  • the segment is transmitted to the terminal main voice system, that is, after detecting that the terminal main voice system is turned off, detecting the voice to determine whether the terminal main voice system needs to be turned on, and if necessary, sending an enable signal to implement the voice-on terminal main voice system.
  • FIG. 1 is a flowchart of a terminal control method according to a first embodiment of the present invention
  • FIG. 2 is a schematic diagram of functional modules of a terminal control device according to a second embodiment of the present invention
  • FIG. 4 is a flowchart of a terminal control method according to a fourth embodiment of the present invention.
  • the core idea of the present invention is to continuously monitor user voice including at least one voice segment when the terminal main voice system is in the off state, and sequentially detect each voice segment in the user voice, when detecting a voice segment with an open attribute.
  • the terminal control method includes the following steps:
  • the terminal main voice system involved in the present invention refers to a voice system in an intelligent device such as an existing mobile phone, such as a siri application in an Apple mobile phone, etc., such a voice system is generally in a closed state when the user is not in use, only in the user.
  • the main operating system of the terminal involved refers to the main operating system of the terminal device such as a mobile phone, such as an IOS system, an Android system, etc., such a main operating system consumes a large amount of power when activated, so when the user does not use the mobile phone, such a type
  • the terminal's main operating system is in a dormant state; detecting whether the terminal's main voice system is turned off.
  • the terminal control device provided by the present invention can detect whether the main voice system is turned off in real time, or first detect whether the main operating system is dormant (the main voice system is definitely turned off when sleeping, and further detecting whether the main voice system is turned off when not sleeping).
  • the main voice system may actively send a signal to inform the terminal control device of the state when the main voice system is off/the main operating system is in sleep; the voice segment involved in the present invention refers to two natural pauses formed when the user speaks.
  • step S102 determining whether a voice segment with an open attribute is present; if yes, executing step S103; if not, proceeding to step S101; in some embodiments, step S102 may be newly added by a DSP chip such as a mobile phone and the like
  • the in-chip program is implemented by the mobile phone CPU and the main operating system.
  • the step S102 may specifically include: sequentially acquiring content feature parameters of each voice segment in the monitored user voice, respectively
  • the content parameter is used to identify the content of the voice segment; if the content feature parameter of a voice segment matches the preset enable parameter, the voice segment has an open attribute; specifically, in this embodiment, the content
  • the feature parameter refers to the feature parameters related to the content of the voice segment, such as the content of the voice segment (the analog-to-digital conversion process is required), the length of the voice segment (duration, no analog-to-digital conversion processing), and the envelope of the voice segment (different After the voice content is uttered, it corresponds to different envelopes, that is, there is a large relationship between the envelope features and the content.
  • the preset opening parameter may be the text content specified/input by the user, or may be a specific length specified by the user (the voice segment satisfying the length may open the terminal)
  • the main voice system naturally, the user can also say a paragraph, the envelope of the paragraph as an open parameter and so on.
  • the matching involved in this embodiment may be set to be identical to reduce the erroneous operation, or a threshold may be set, and the voice segment having the similarity greater than the threshold may be used as the voice segment that satisfies the condition to enable the terminal main voice system, which is compatible.
  • the pronunciation length of the same control voice "Hello Little 5" is different, such as the time spent by the user in the morning (the user's brain is not awake, the response is slow) It may be 1.2 seconds, and at noon (the user's brain is awake, the response is fast), the time taken by the user may be 0.95 seconds. It is assumed that the length of the voice segment in the preset open parameter set by the user is 1 second and the threshold is 90%.
  • the method further includes: acquiring the identification parameters of each voice segment, and respectively comparing with the preset legal user identification parameters, where the identification parameter is used. Identifying the speaker identity of the voice segment; only when the content feature parameter matches, and the identity parameter compares the legal voice segment, the open signal is sent to the terminal main voice system, and is stored after being monitored in the voice segment with the open attribute.
  • the identification parameter may be a self-characteristic parameter that is related to the speaker's pronunciation and is independent of the voice content, such as voiceprints and frequencies of the voice segment.
  • the default legal user identification parameter is the terminal owner/legal user said Voiceprint, frequency, etc. in the speech; in this embodiment, the speaker identity of the voice segment is identified to prevent the terminal from being illegally used; and the matching of the content feature parameters and the identity parameter parameters that may be involved in this step are not in the order, Select one or two parameters as needed to detect the speech segment conditions.
  • S103 Sending an open signal for enabling the main voice system of the terminal, and storing the voice segment monitored after the voice segment with the open attribute; the sending start signal involved in this step and storing the subsequent voice segment are two parallel processes. In this way, the problem that the loss of the voice segment or the incomplete reception of the voice segment is incomplete during the process of the terminal main voice system being turned on is avoided; S104: After detecting that the terminal voice system is successfully turned on, the stored voice segment is transmitted; In some embodiments, the function of detecting whether the terminal main voice system is successfully turned on corresponding to detecting whether the main voice system is turned off may be that the terminal control apparatus provided by the present invention sends an enable signal after the main operating system is not sleeping.
  • the sent start signal also has the function of waking up the main operating system, and after detecting the start signal, detecting whether the main operating system wakes up in real time, and if waking up, further Detect whether the main voice system is turned on; corresponding, the The function may be that after the primary voice system is turned on, the active voice signal is sent to inform the terminal control device that the device has been successfully turned on, or the master voice system/main operating system actively detects whether the terminal control device stores the voice segment after the start/wake-up, when the terminal After receiving the corresponding command, the control device considers that the main voice system is successfully turned on.
  • the method before performing the segment of the voice segment stored in the step S103 after the voice segment having the open attribute, the method further includes: storing a voice segment having an open attribute; thus avoiding when a voice segment is enabled. When the attribute has control attribute, the control instruction is lost.
  • the control attribute means that the terminal needs to complete an operation according to a certain voice segment.
  • the voice segment of "small 5 check weather” has both the opening attribute (turning on the main voice system), There is also a control attribute (query weather). If it is only used as an open command, it will result in the loss of the control command.
  • the foregoing embodiment further includes: determining whether the voice segment having the open attribute has a control attribute, and if yes, storing the voice segment having the open attribute; In this embodiment, before storing the voice segment with the open attribute, it is determined whether the control attribute is provided at the same time, the storage of the voice segment without the control attribute is avoided, and the data storage amount is reduced.
  • the foregoing embodiment further includes: after detecting that the terminal main voice system is successfully turned on, determining whether the voice segment is being stored, and if yes, continuing to store the voice segment, and after the voice segment is interrupted, terminating the subsequent voice
  • the storage of the segment if not, terminates the storage of the subsequent voice segment; after terminating the storage of the subsequent voice segment, the subsequent voice segment and all stored voice segments are transmitted to the terminal primary voice system; for example, the user voice to be said by the user is "Hello little 5, help me check the traffic, I want to eat.”
  • the terminal's main voice system turns on the success signal, the "help me check traffic" voice segment (such as storage to "check”), then , continue to store "help me check traffic", and after the end of the storage, terminate the storage of the subsequent voice segment "I want to eat" (by the terminal voice system that has been opened, such as siri receiving processing), if the terminal is in the main voice system
  • the success signal is turned on, the storage of "Help me check traffic" has
  • the subsequent speech segment will be terminated directly.
  • I want to eat "storage” (received by the terminal main voice system that has been turned on, such as siri reception processing); whether the voice segment of the terminal main voice system when the terminal voice system is turned on or not is given a specific scheme.
  • the terminal primary voice system sequentially processes or sequentially transmits to the network side to process multiple voices according to the listening time of each voice segment.
  • the user needs to call *** to eat together, the user voice is "Hello little 5, please call ***, 6 pm, eat at a certain restaurant", “Hello little in the user's voice” 5 "Used to enable the terminal main voice system, when the "Hello 5" is detected, the main voice system is turned on, and the memory "Please call ***, PM” is stored, when stored to "PM", the main The voice system is successfully turned on.
  • the judgment result is that the voice segment of "6 pm” is being stored, and the storage is continued.
  • the subsequent voice segment "eat at a certain restaurant” is received by the siri of the main voice system, and the user finishes the user voice.
  • FIG. 2 is a functional block diagram of a terminal control device according to a second embodiment of the present invention. As shown in FIG.
  • the terminal control device 2 provided by the present invention includes the following functional modules: After detecting that the terminal main voice system is turned off, listening to the voice monitoring module 21 of the user voice including at least one voice segment; An attribute determining module 22 configured to determine whether a voice segment having an open attribute is present; if it is set, sending an open signal for turning on the terminal main voice system, and storing the voice segment monitored after the voice segment having the open attribute The storage module 23 is opened; and the data transmission module 24 is configured to transmit the stored voice segment after detecting that the terminal main voice system is successfully turned on. In some embodiments, the data transmission module 24 in the embodiment shown in FIG.
  • the voice control device 3 includes: setting to determine whether a voice segment is being stored after detecting that the terminal primary voice system is successfully turned on, and if so, continuing to store the voice segment, and Terminating the storage of the subsequent voice segment after the interruption of the voice segment, if not, terminating the storage control module for storing the subsequent voice segment; setting to transmit the subsequent voice segment and all stored voice segments after terminating the storage of the subsequent voice segment Outgoing transmission control module.
  • 3 is a schematic diagram of a function module of a voice control device according to a third embodiment of the present invention. As shown in FIG. 3, in the embodiment, the voice control device 3 provided by the present invention includes a monitoring module 31 and a detection module.
  • the control module 33 and the storage module 34 are configured to monitor the user voice including at least one voice segment after detecting that the terminal main voice system is turned off, and transmit the voice to the detection module 32.
  • the monitoring module 31 It may be a device/structure with a voice receiving function, such as a microphone of the terminal device, which can also monitor the external sound after the terminal detects that the terminal main voice system is turned off; the detecting module 32 is configured to determine the user voice monitored by the monitoring module 31.
  • the control module 33 is configured to send an open signal for turning on the terminal main voice system when triggered by the detection module 32, and
  • the control storage module 34 stores the voice segment that is subsequently monitored by the processing module, and is also set to detect After the terminal main voice system is successfully turned on, the control storage module 34 transmits the stored voice segment, which is generally transmitted to the terminal main voice system for processing. If the terminal primary voice system cannot be processed by itself, it needs to be transmitted to the network side for processing; Module 34 is arranged to store or transmit voice segments under the control of control module 33.
  • the detecting module 32 in the embodiment shown in FIG.
  • the detecting module 32 in the embodiment shown in FIG. 3 is further configured to acquire the identification parameters of each voice segment before the triggering control module 33, and respectively compare with the preset legal user identification parameters, the identity.
  • the identification parameter is used to identify the speaker identity of the voice segment; the control module 33 is triggered only when the content feature parameter matches and the identity parameter matches the legal voice segment.
  • control module 33 in the embodiment shown in FIG. 3 is further configured to control the storage module 34 to store a voice segment having an open attribute before storing the voice segment that is monitored after the voice segment having the open attribute.
  • control module 33 in the embodiment shown in FIG. 3 is further configured to control whether the voice segment having the open attribute has a control attribute before the storage module 34 stores the voice segment having the open attribute, and if so, control storage.
  • Module 34 stores a speech segment with an open attribute.
  • the control module 33 in the embodiment shown in FIG. 3 is further configured to: after detecting that the terminal main voice system is successfully turned on, determine whether the storage module is storing any voice segment; if yes, the control storage module continues to store the voice segment.
  • the present invention also provides a terminal for using the control technology provided by the present invention, the terminal comprising a terminal main voice system and a voice control device 3 as shown in FIG. 3; the voice control device 3 is arranged to detect the terminal After the main voice system is turned off, the user voice including at least one voice segment is monitored to determine whether a voice segment with an open attribute is present.
  • the terminal main voice system When a voice segment with an open attribute appears, the terminal main voice system is turned on, and the voice with the open attribute is stored.
  • the voice segment monitored after the segment, and after the terminal main voice system is turned on, the stored voice segment is handed over to the terminal main voice system.
  • the voice segment that the terminal primary voice system needs to process has two sources: a voice segment stored by the voice control device and a subsequent voice segment received after the terminal primary voice system is turned on, so if the subsequent voice segment and/or
  • the stored voice segment is a plurality of voice segments, and the terminal primary voice system is configured to process multiple voice segments in sequence or sequentially transmit multiple voice segments to the network side according to the listening time sequence of each voice segment.
  • FIG. 4 is a flowchart of the terminal control method according to the fourth embodiment of the present invention.
  • the terminal control method provided by the present invention includes the following steps: S401: The user does not operate the mobile phone for a long time, and the mobile phone main operating system enters a sleep state; if the user does not operate the mobile phone for 5 minutes, the mobile phone automatically enters a sleep state, and the main voice system At the same time, it enters the closed state, sends a shutdown signal to the terminal control device, and the user places the mobile phone next to the computer;
  • the terminal control device monitors each voice segment in the user voice; the user uses the QQ video to talk to the small A, and the assumed dialogue occurs, and the terminal control device monitors the conversation;
  • step S403 The terminal control device determines, for each voice segment, whether it has an open attribute, if yes, step S404 is performed, otherwise, returns to step S402; the terminal performs D/A conversion on each voice segment that is monitored to obtain a voice segment.
  • the default opening parameter is "Hello Small 5" (personalized naming, you can avoid the hidden danger of the general command "Hello mobile phone” that is inadvertently opened by other users), the voice segment “small A waits", "I If the result of the notification "small B dinner at night” is no, the process returns to step S402, and the result of the speech segment "Hello small 5" is YES, step S404 is performed;
  • step S404 The terminal control device determines the voice segment having the open attribute, and determines whether it is legal. If yes, the process goes to step S405. If no, the process returns to step S402.
  • the voice segment of the voice segment "Hello Small 5" is determined and the default legal user is determined. Whether the voiceprint of the (mobile phone holder) is consistent. If they are consistent, the voice segment is legal. Otherwise, it is illegal.
  • the terminal control device sends an open signal, and starts storing the subsequent voice segment; sends an open signal to the terminal primary voice system (the open signal is used to wake up the main operating system IOS and the voice system siri), and starts to "call the phone small B.” , ... the storage of ;
  • step S406 The terminal control device receives the start success signal, and detects whether a certain voice segment is being stored. If yes, step S407 is performed. If no, step S408 is performed; if the siri voice system is successfully turned on, the sending is started. When the signal is successful, the "seven o'clock" speech segment is being stored, step S407 is performed, and if not, step S408 is performed; S407: The terminal control device continues to store the voice segment; and after the storage of the voice segment is completed, step S408 is performed; and the assumption of step S406 is followed, after the storage at "6 pm" is completed (the voice interruption is greater than 0.3 seconds), and execution is performed.
  • Step S408 S408: The terminal control device terminates the storage of the subsequent voice segment, and the subsequent voice segment is received by the primary voice system, and the stored voice segment is transmitted to the main voice system.
  • the assumption of step S407 is followed, and the subsequent voice segment is "goed.”"Some restaurant eating" is received by the main voice system siri (can also be transmitted by the main operating system IOS and then transmitted to siri processing); at the same time, the voice segment "please call small B" and "night 6" are transmitted to the terminal main voice system;
  • S409 The main voice system sequentially processes the voice segment in the order of the listening time of each voice segment, and displays the voice segment to the user; the voice segment received by the main voice system is chronologically followed by "please call small B", "night at 6 o'clock" and " Going to a certain restaurant to eat", the main voice system is processed in turn with the cooperation of the network side: For the "Please call the small B", the terminal main voice system calls the address book, and to the small B For the
  • This embodiment implements the functions of the automatic opening of the main voice system, the continuous storage of the voice segment, and the like, and only an exemplary application scenario of the present invention is given.
  • the timing of performing step S406 in this application scenario is the success of the main voice system.
  • the timing of executing step S406 is after the main operating system wakes up successfully, because the wakeup of the main operating system is before the main voice system is turned on, so that after the main operating system successfully wakes up, the main operation is performed.
  • the system performs the subsequent storage of the voice segment, and the effect is better (the main operating system receives the stored voice segment to reduce the storage requirement for the DSP chip).
  • the implementation of the present invention has at least the following beneficial effects: when the terminal main voice system is in the off state, continuously listening to the user voice including at least one voice segment, and sequentially detecting each voice segment in the user voice.
  • a voice segment with an open attribute is detected, an open signal is sent to the terminal main voice system, and at the same time, the voice segment that is stored after the voice segment with the open attribute is started, and the stored voice is awake after the terminal main voice system wakes up.
  • the segment is transmitted out, that is, after detecting that the terminal main voice system is turned off, detecting the voice to determine whether the terminal primary voice system needs to be turned on, and if necessary, sending The signal is turned on, and the main voice system of the voice-on terminal is realized.
  • the subsequent voice segments are stored in parallel, and the user does not need to wait for the main voice system of the terminal to be turned on, and the voice command can be continuously spoken without worrying about the voice command terminal.
  • the voice system is not available, which solves the complicated operation of the voice control technology in the prior art and enhances the user experience.
  • the above is only a specific embodiment of the present invention, and is not intended to limit the present invention in any way. Any simple modification, equivalent change, combination or modification of the above embodiments in accordance with the technical spirit of the present invention is still in the present invention. The scope of protection of the technical solution of the invention.
  • the above technical solution provided by the present invention can be applied to a terminal control process, and when a terminal main voice system is in a closed state, continuously listening for user voice including at least one voice segment, for each voice segment in the user voice Detecting in turn, when detecting a voice segment with an open attribute, sending an open signal to the terminal main voice system, and simultaneously starting the voice segment that is stored after the voice segment having the open attribute, after the terminal main voice system wakes up,
  • the technical solution of transmitting the stored voice segment to the terminal main voice system the user does not need to wait for the terminal main voice system to be turned on, can continuously speak the voice command, and does not need to worry about the voice command terminal main voice system cannot be obtained, and solves the prior art.
  • the problem of complicated operation of voice control technology enhances the user experience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Computational Linguistics (AREA)
  • Signal Processing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Telephone Function (AREA)
  • Telephonic Communication Services (AREA)

Abstract

本发明提供了一种终端控制方法、装置、语音控制装置及终端，该方法包括：在检测到终端主语音系统关闭后，监听包括至少一个语音段的用户语音，判断是否出现具备开启属性的语音段，若出现，则发送用于开启终端主语音系统的开启信号，并存储在具备开启属性的语音段之后监听到的语音段，在检测到终端主语音系统成功开启后，将存储的语音段传输出去。通过本发明的实施，在终端主语音系统关闭时，检测语音段来判断是否需要开启终端主语音系统，若需要，则发送开启信号，同时，在发送开启信号后，并行存储后续语音段落，用户不需要等待终端主语音系统开启，即可连续说出语音指令，解决了现有技术中语音控制技术操作繁杂的问题。

Description

一种终端控制方法、 装置、 语音控制装置及终端 技术领域 本发明涉及终端控制领域， 尤其涉及一种终端控制方法、 装置、 语音控制装置及 终端。 背景技术 随着智能终端处理能力的增强,用户对智能终端也提出了越来越高的要求,如为了 解放用户的双手,现有技术提供了多种多样的语音识别及控制技术，诸如百度谷歌的语 音输入、科大讯飞的语音识别及 apple的 siri等语音系统， 这些应用给用户带来较好的 便利性， 解放双手无需打字输入， 也带来很大的娱乐性， 其本地的或基于互联网云计 算的语音智能识别可以给用户提供互动式交互。 现有的语音系统在提高了用户使用体验的同时， 还存在这样的问题： 用户首先需 要按住语音按钮， 进入语音系统， 然后才能发出语音指令， 当然， 若手机处于关闭状 态， 则需要先唤醒手机， 在按语音按钮进行语音系统； 如用户在手机关闭的时候， 需 要查询一下天气， 使用现有的产品需要先拿起手机并解锁， 激活主操作系统， 在按语 音按钮，开启语音系统（如 siri ),在语音系统开启后才能发出语音指令"查一下天气"， 步骤比较繁琐。 因此， 如何提供一种可以降低用户操作繁杂度的终端控制技术， 是本领域技术人 员亟待解决的技术问题。 发明内容 本发明实施例提供了一种终端控制方法、 装置、 语音控制装置及终端， 以解决现 有技术中语音控制技术操作繁杂的问题。 本发明实施例提供了一种终端控制方法，在一个实施例中，该终端控制方法包括： 在检测到终端主语音系统关闭后， 监听包括至少一个语音段的用户语音， 判断是否出 现具备开启属性的语音段， 若出现， 则发送用于开启终端主语音系统的开启信号， 并 存储在具备开启属性的语音段之后监听到的语音段， 在检测到终端主语音系统成功开 启后， 将存储的语音段传输出去。 优选地， 上述实施例中的判断是否出现具备开启属性的语音段的步骤包括： 依次 获取监听到的用户语音中各语音段的内容特征参数， 分别与预设开启参数匹配,内容特 征参数用于识别语音段的内容；若一语音段的内容特征参数与预设开启参数匹配通过， 则该语音段具备开启属性。 优选地， 上述实施例在发送开启信号之前还包括： 获取各语音段的身份识别参数， 并分别与预设合法用户身份识别参数比对,身份识别参数用于识别语音段的说话人身 份； 仅在出现内容特征参数匹配通过、 身份识别参数比对合法的语音段时， 发送开启 信号， 并存储在具备开启属性的语音段之后监听到的语音段。 优选地， 上述实施例在存储在具备开启属性的语音段之后监听到的语音段之前， 还包括： 存储具备开启属性的语音段。 优选地， 上述实施例在存储具备开启属性的语音段之前， 还包括： 判断具备开启 属性的语音段是否同时具备控制属性， 若是， 则存储具备开启属性的语音段。 优选地， 上述实施例还包括： 在检测到终端主语音系统成功开启后， 判断是否正 在存储语音段； 若是， 则继续存储该语音段， 并在该语音段中断后， 终止后续语音段 的存储， 若否， 则终止后续语音段的存储； 在终止后续语音段的存储后， 将后续语音 段及存储的所有语音段传输出去。 优选地， 在上述实施例中， 若后续语音段和 /或存储的所有语音段为多个语音段， 则终端主语音系统按照各语音段的监听时间先后， 依次处理多个语音段或依次传输多 个语音段至网络侧。 本发明实施例还提供了一种终端控制装置， 在一个实施例中， 该终端控制装置包 括： 设置为在检测到终端主语音系统关闭后， 监听包括至少一个语音段的用户语音的 语音监听模块； 设置为判断是否出现具备开启属性的语音段的属性判断模块； 设置为 若出现， 则发送用于开启终端主语音系统的开启信号， 并存储在具备开启属性的语音 段之后监听到的语音段的开启存储模块； 以及设置为在检测到终端主语音系统成功开 启后， 将存储的语音段传输出去的数据传输模块。 为将本发明实施例运用于实际生活中，本发明实施例也提供了一种语音控制装置， 在一个实施例中， 该语音控制装置包括监听模块、 检测模块、 控制模块及存储模块， 其中， 监听模块设置为在检测到终端主语音系统关闭后， 监听包括至少一个语音段的 用户语音， 并传输至检测模块； 检测模块设置为判断监听模块监听到的用户语音中是 否出现具备开启属性的语音段， 并在出现具备开启属性的语音段时， 触发控制模块； 控制模块设置为在被检测模块触发时发送用于开启终端主语音系统的开启信号， 并控 制存储模块存储监听模块在具备开启属性的语音段之后监听到的语音段， 还设置为在 检测到终端主语音系统成功开启后， 控制存储模块将存储的语音段传输出去； 存储模 块设置为在控制模块的控制下存储或传输语音段。 同时， 本发明实施例也提供了一种终端， 在一个实施例中， 该终端包括终端主语 音系统及本发明提供的语音控制装置；语音控制装置设置为在终端主语音系统关闭时， 监听包括至少一个语音段的用户语音， 判断是否出现具备开启属性的语音段， 在出现 具备开启属性的语音段时， 开启终端主语音系统， 并存储在具备开启属性的语音段之 后监听到的语音段， 并在终端主语音系统开启后， 将存储的语音段交由终端主语音系 统。 本发明实施例的有益效果： 本发明实施例提供的方法， 在终端主语音系统处于关闭状态时， 持续监听包括至 少一个语音段的用户语音， 针对用户语音中的每个语音段依次进行检测， 当检测到具 备开启属性的语音段时， 向终端主语音系统发送开启信号， 并同时开始存储在具备开 启属性的语音段之后监听到的语音段， 在终端主语音系统唤醒后， 将存储的语音段传 输至终端主语音系统， 也即， 在检测到终端主语音系统关闭后， 检测语音来判断是否 需要开启终端主语音系统， 若需要， 则发送开启信号， 实现了语音开启终端主语音系 统， 同时， 在发送开启信号后， 并行存储后续语音段， 用户不需要等待终端主语音系 统开启， 即可连续说出语音指令， 也无需担心语音指令终端主语音系统无法获得， 解 决了现有技术中语音控制技术操作繁杂的问题， 增强了用户的使用体验。 附图说明 图 1为本发明第一实施例提供的终端控制方法的流程图； 图 2为本发明第二实施例提供的终端控制装置的功能模块示意图； 图 3为本发明第三实施例提供的语音控制装置的功能模块示意图； 图 4为本发明第四实施例提供的终端控制方法的流程图。 具体实施方式 现通过具体实施方式结合附图的方式对本发明做出进一步的诠释说明。 本发明的核心思想是在终端主语音系统处于关闭状态时， 持续监听包括至少一个 语音段的用户语音， 针对用户语音中的每个语音段依次进行检测， 当检测到具备开启 属性的语音段时， 向终端主语音系统发送开启信号， 并同时开始存储在具备开启属性 的语音段之后监听到的语音段， 在终端主语音系统唤醒后， 将存储的语音段传输至终 端主语音系统， 也即， 在检测到终端主语音系统关闭后， 检测语音来判断是否需要开 启终端主语音系统， 若需要， 则发送开启信号， 实现了语音开启终端主语音系统， 同 时， 在发送开启信号后， 并行存储后续语音段， 用户不需要等待终端主语音系统开启， 即可连续说出语音指令， 也无需担心语音指令终端主语音系统无法获得。 第一实施例： 图 1为本发明第一实施例提供的终端控制方法的流程图， 由图 1可知， 在本实施 例中， 本发明提供的终端控制方法包括以下步骤：
S101 : 在检测到终端主语音系统关闭后， 监听包括至少一个语音段的用户语音； 本步骤可以通过终端内的语音采集装置， 如麦克风来采集， 对采集到的每个语音 段均执行步骤 S102; 本发明所涉及的终端主语音系统是指现有手机等智能设备中的语音系统， 如苹果 手机内的 siri应用等， 这类语音系统在用户不使用时， 一般处于关闭状态， 仅在用户 需要时， 通过按语音按钮等方式来开始； 主语音系统关闭包括两种情况： 终端主操作 系统未休眠时， 退出了语音应用， 以及主操作系统休眠时， 语音应用肯定是关闭的， 本发明所涉及的终端主操作系统是指手机等终端设备的主要操作系统， 如 IOS系统、 安卓系统等， 这类主操作系统在激活状态时耗电量大， 因此在用户未使用手机时， 这 类终端主操作系统处于休眠状态； 检测终端主语音系统是否关闭这一功能可以是本发明所提供的终端控制装置来实 时检测主语音系统是否关闭， 或者先检测主操作系统是否休眠 （休眠时， 主语音系统 肯定关闭， 在未休眠时进一步检测主语音系统是否关闭）， 对应的， 还可以是主语音系 统在关闭时 /主操作系统在休眠时， 主动发送信号告知终端控制装置其状态； 本发明所涉及的语音段是指用户在说话时两个自然停顿所形成的段落， 如用户说 一句话 "你好小 5， 请查询天气"， 此时 "你好小 5 "与 "请查询天气"之间的自然停 顿大于各个字节之间的时间间隔， 这个时间间隔是终端可以分辨识别的， 本发明不做 过多说明， 而 "你好小 5 "与 "请查询天气"则是两个语音段； 而用户语音则是用户 所说的至少一个语音段连接形成的； S102: 判断是否出现具备开启属性的语音段； 若出现， 则执行步骤 S103 , 若未出 现， 则继续执行步骤 S101 ; 在一些实施例中， 步骤 S102可以由手机等终端内新增加的 DSP芯片及芯片内程 序来实现， 也可以由手机 CPU及主操作系统来实现； 在一些实施例中，步骤 S102具体的可以包括：依次获取监听到的用户语音中各语 音段的内容特征参数， 分别与预设开启参数匹配,内容特征参数用于识别语音段的内 容； 若一语音段的内容特征参数与预设开启参数匹配通过，则该语音段具备开启属性； 具体的， 在本实施例中， 内容特征参数是指与语音段的内容相关的特征参数， 如语音 段的内容（需要进行模数转换处理）、语音段的长度（持续时间， 无需模数转换处理）、 语音段的包络 （不同的语音内容在发声后， 就对应着不同的包络， 也即包络特征与内 容之间存在较大的对应关系， 而获取包络也需要模数转换处理） 等参数， 预设开启参 数可以是用户指定 /输入的文字内容， 也可以是用户指定的特定长度（满足该长度的语 音段均可开启终端主语音系统）， 自然也可以是用户说一段话，将这段话的包络作为开 启参数等等。 本实施例所涉及的匹配可以是设置为完全相同以降低误操作， 也可以设 置一个阈值， 相似度大于该阈值的语音段都可以作为满足条件的语音段来开启终端主 语音系统， 这样可以兼容不同应用环境； 例如， 用户处于不同状态 /场景时， 对同一控 制语音 "你好小 5 " 的发音长度不尽相同， 如在早上 （用户大脑不清醒， 反应较慢） 时， 用户所用的时间可能为 1.2秒， 而在中午 （用户大脑清醒， 反应快） 时， 用户所 用的时间可能为 0.95秒， 假定用户设置的预设开启参数中的语音段的长度为 1秒、 阈 值为 90%，那么用户在早上说出"你好小 5 "这一语音段的长度相似度为（1.2-1 )/1=80%， 显然是小于阈值的， 则判定为用户在早上说出 "你好小 5 "不具备开启属性， 同样的， 那么用户在中午说出 "你好小 5 "这一语音段的长度相似度为 0.95/1=95%， 显然是大 于阈值的， 则判定为用户在中午说出 "你好小 5 "具备开启属性； 本实施例一语音段 的长度为例进行了示例性说明，语音段的包络、 内容等同样可以采用这类的匹配机制； 在一些实施例中， 为了避免终端被非法用户使用， 在开启终端主语音系统之前还 包括： 获取各语音段的身份识别参数， 并分别与预设合法用户身份识别参数比对，身份 识别参数用于识别语音段的说话人身份； 仅在出现内容特征参数匹配通过、 身份识别 参数比对合法的语音段时， 发送开启信号至终端主语音系统， 并存储在具备开启属性 的语音段之后监听到的语音段； 具体的， 在本实施例中， 身份识别参数可以是语音段 的声纹、 频率等语音段所表现出来的仅与说话人的发音相关而与语音内容无关的自身 特征参数， 那么， 对应的， 预设合法用户身份识别参数就是终端拥有者 /合法使用者说 话时的声纹、 频率等； 本实施例对语音段的说话人身份进行识别， 避免终端被非法使 用； 而本步骤可能涉及的内容特征参数匹配与身份识别参数比对并没有先后顺序， 可 以根据需要来选择一个或两个参数来实现语音段条件的检测。 S103 : 发送用于开启终端主语音系统的开启信号， 并存储在具备开启属性的语音 段之后监听到的语音段； 本步骤所涉及的发送开启信号及存储后续语音段是两个并行的过程， 这样， 就避 免了在终端主语音系统开启的过程中语音段的丢失或者语音段接收不完整等问题的出 现； S104: 在检测到终端主语音系统成功开启后， 将存储的语音段传输出去； 在一些实施例中， 与检测主语音系统是否关闭相对应的， 检测终端主语音系统是 否成功开启这一功能可以是本发明所提供的终端控制装置在主操作系统未休眠时， 发 送开启信号后， 实时检测主语音系统是否开启成功， 或者在主操作系统休眠时， 所发 送的开启信号还同时具备唤醒主操作系统功能， 在发送开启信号后， 实时检测主操作 系统是否唤醒， 若唤醒则进一步检测主语音系统是否开启； 对应的， 该功能可以是主 语音系统在开启后， 主动发送信号告知终端控制装置其已经开启成功， 或者是主语音 系统 /主操作系统在开启 /唤醒后，主动检测终端控制装置是否存储了语音段， 当终端控 制装置接收到对应指令后， 认为主语音系统开启成功。 本实施例同时执行开启终端主语音系统及存储后续语音段两个操作， 避免了在终 端主语音系统开启的过程中语音段的丢失或者语音段接收不完整， 这样， 终端主语音 系统在开启后就可以得到完整的语音段进行处理， 不会出现因错误语音段导致的执行 结果错误等问题。 在一些实施例中，在执行步骤 S103中的存储在具备开启属性的语音段之后监听到 的语音段之前， 还包括： 存储具备开启属性的语音段； 这样就避免了当某语音段既有 开启属性、 又有控制属性时， 控制指令的丢失； 控制属性是指终端需要根据某语音段 完成一个操作， 如 "小 5查天气"这一语音段就既有开启属性（开启主语音系统）、 又 有控制属性 （查询天气）， 若仅作为开启指令， 就将导致控制指令的丢失。 在一些实施例中， 上述实施例在存储具备开启属性的语音段之前， 还包括： 判断 具备开启属性的语音段是否同时具备控制属性，若是，则存储具备开启属性的语音段; 本实施例在存储具备开启属性的语音段之前， 判断是否同时具备控制属性， 避免了不 具备控制属性的语音段的存储， 降低了数据存储量。 在一些实施例中， 上述实施例还包括： 在检测到终端主语音系统成功开启后， 判 断是否正在存储语音段， 若是， 则继续存储该语音段， 并在该语音段中断后， 终止后 续语音段的存储， 若否， 则终止后续语音段的存储； 在终止后续语音段的存储后， 将 后续语音段及存储的所有语音段传输至终端主语音系统； 例如， 用户所要说的用户语 音为 "你好小 5， 帮我查一下交通， 我要吃饭"若在终端主语音系统开启成功信号时， 正在存储 "帮我查一下交通"这一语音段 （如存储到 "查" )， 那么， 继续存储 "帮我 查一下交通"， 并在存储结束后， 终止后续语音段"我要吃饭"的存储（由已经开启的 终端主语音系统、 如 siri接收处理）， 若在终端主语音系统开启成功信号时， 已经完成 "帮我查一下交通" 的存储， 还未开始 "我要吃饭" 的存储， 则直接终止后续语音段 "我要吃饭"的存储 （由已经开启的终端主语音系统、 如 siri接收处理）； 本实施例终 端主语音系统开启时 /后的语音段是否进行存储给出了具体的方案。 在一些实施例中， 若后续语音段和 /或存储的所有语音段为多个语音段， 则终端主 语音系统按照各语音段的监听时间先后， 依次处理或依次传输至网络侧处理多个语音 段； 如， 用户需要打电话给 ***一起吃饭， 用户语音为 "你好小 5， 请电话 ***， 下午 6 点， 在某某餐厅吃饭"， 该用户语音中的 "你好小 5 "用来开启终端主语音系统， 在检 测到 "你好小 5 " 时， 发送主语音系统开启信号， 并开始存储 "请电话 ***， 下午"， 当存储到 "下午" 时， 主语音系统开启成功， 此时判断结果为正在存储 "下午 6点" 这一语音段， 在继续存储， 后续语音段 "在某某餐厅吃饭"则由主语音系统的 siri接 收， 用户说完用户语音后， 后续语音段和 /或存储的所有语音段为 "请电话 ***， 下午 6 点， 在某某餐厅吃饭"这三个语音段， 终端主语音系统先执行电话操作 "请电话 *** "， 然后制定日程安排 "下午 6点， 某某餐厅在吃饭"， 而 "在某某餐厅吃饭"则需要到网 络订餐、查询交通路线等相互配合， 终端主语音系统无法独立完成， 需要网络侧处理； 这样， 终端主语音系统或网络侧就可以按照时间先后顺序来及时向用户反馈各语音指 令的处理结果。 第二实施例： 图 2为本发明第二实施例提供的终端控制装置的功能模块示意图， 由图 2可知， 在本实施例中， 本发明提供的终端控制装置 2包括以下功能模块： 设置为在检测到终端主语音系统关闭后， 监听包括至少一个语音段的用户语音的 语音监听模块 21 ; 设置为判断是否出现具备开启属性的语音段的属性判断模块 22; 设置为若出现， 则发送用于开启终端主语音系统的开启信号， 并存储在具备开启 属性的语音段之后监听到的语音段的开启存储模块 23 ; 以及 设置为在检测到终端主语音系统成功开启后， 将存储的语音段传输出去的数据传 输模块 24。 在一些实施例中， 图 2所示实施例中的数据传输模块 24包括： 设置为在检测到终 端主语音系统成功开启后， 判断是否正在存储语音段， 若是， 则继续存储该语音段， 并在该语音段中断后终止后续语音段的存储， 若否， 则终止后续语音段的存储的存储 控制模块； 设置为在终止后续语音段的存储后， 将后续语音段及存储的所有语音段传 输出去的传输控制模块。 第三实施例： 图 3为本发明第三实施例提供的语音控制装置的功能模块示意图， 由图 3可知， 在本实施例中， 本发明提供的语音控制装置 3包括监听模块 31、检测模块 32、控制模 块 33及存储模块 34， 其中， 监听模块 31设置为在检测到终端主语音系统关闭后，监听包括至少一个语音段的 用户语音， 并传输至检测模块 32; 具体的， 监听模块 31可以是终端设备的麦克风等 具备语音接收功能的装置 /结构， 其在终端在检测到终端主语音系统关闭后， 也可以监 听外界声音； 检测模块 32设置为判断监听模块 31监听到的用户语音中是否出现具备开启属性 的语音段， 并在出现具备开启属性的语音段时， 触发控制模块 33 ; 控制模块 33设置为在被检测模块 32触发时发送用于开启终端主语音系统的开启 信号， 并控制存储模块 34存储处理模块后续监听到的语音段，还设置为在检测到终端 主语音系统成功开启后，控制存储模块 34将存储的语音段传输出去，一般为传输至终 端主语音系统处理，若终端主语音系统不能独自处理，还需要传输至网络侧进行处理； 存储模块 34设置为在控制模块 33的控制下存储或传输语音段。 在一些实施例中， 图 3所示实施例中的检测模块 32设置为： 依次获取监听到的用 户语音中各语音段的内容特征参数， 分别与预设开启参数匹配,内容特征参数用于识别 语音段的内容； 若一语音段的内容特征参数与预设开启参数匹配通过， 则该语音段具 备开启属性。 在一些实施例中， 图 3所示实施例中的检测模块 32还设置为在触发控制模块 33 之前， 获取各语音段的身份识别参数， 并分别与预设合法用户身份识别参数比对，身份 识别参数用于识别语音段的说话人身份； 仅在出现内容特征参数匹配通过、 身份识别 参数比对合法的语音段时， 触发控制模块 33。 在一些实施例中， 图 3所示实施例中控制模块 33还设置为控制存储模块 34在存 储在具备开启属性的语音段之后监听到的语音段之前， 存储具备开启属性的语音段。 在一些实施例中， 图 3所示实施例中控制模块 33还设置为控制存储模块 34存储 具备开启属性的语音段之前， 判断具备开启属性的语音段是否同时具备控制属性， 若 是， 则控制存储模块 34存储具备开启属性的语音段。 在一些实施例中，图 3所示实施例中控制模块 33还设置为在检测到终端主语音系 统成功开启后， 判断存储模块是否正在存储任一语音段； 若是， 则控制存储模块继续 存储该语音段， 并在该语音段中断后终止后续语音段的存储， 若否， 则控制存储模块 终止后续语音段的存储； 在终止后续语音段的存储后， 将后续语音段传输至终端主语 音系统， 并控制存储模块将存储的所有语音段传输至终端主语音系统。 在一些实施例中， 本发明还提供了终端来使用本发明提供的控制技术， 该终端包 括终端主语音系统及如图 3所示的语音控制装置 3 ; 语音控制装置 3设置为在检测到 终端主语音系统关闭后， 监听包括至少一个语音段的用户语音， 判断是否出现具备开 启属性的语音段， 在出现具备开启属性的语音段时， 开启终端主语音系统， 并存储在 具备开启属性的语音段之后监听到的语音段， 并在终端主语音系统开启后， 将存储的 语音段交由终端主语音系统。 在一些实施例中， 终端主语音系统需要处理的语音段有两个来源： 语音控制装置 存储的语音段及终端主语音系统开启后接收到的后续语音段， 因此， 若后续语音段和 / 或存储的所有语音段为多个语音段， 则终端主语音系统设置为按照各语音段的监听时 间先后， 依次处理多个语音段或依次传输多个语音段至网络侧。 第四实施例： 现结合一具体应用实例来说明本发明的主要技术方案， 在本应用实例中， 假定终 端为苹果 5手机（主语音系统为 IOS)， 用户所说的语音为 "……， 小 A等一下， 我通 知小 B晚上一起吃饭； 你好小 5， 请电话小 B， 晚上 6点， 去某某餐厅吃饭。"终端判 断某语音段是否具备开启属性的依据是语音段内容， 并利用声纹检测用户是否合法； 图 4为本发明第四实施例提供的终端控制方法的流程图， 由图 4可知，在本实施例中， 本发明提供的终端控制方法包括以下步骤： S401 : 用户长时间未操作手机， 手机主操作系统进入休眠状态； 如用户 5分钟没有操作手机， 手机自动进入休眠状态， 主语音系统同时进入关闭 状态， 发送关闭信号到终端控制装置， 用户把手机放在电脑旁边；
S402: 终端控制装置监听用户语音中的每一个语音段； 用户利用 QQ视频与小 A通话， 发生上述假定的对话， 终端控制装置监听到上述 对话；
S403 : 终端控制装置对每一个语音段进行判断， 判断是否具备开启属性， 是， 则 执行步骤 S404, 否， 则返回步骤 S402; 终端对监听的每一个语音段都进行 D/A转换获取语音段内容，预设开启参数为"你 好小 5 " (个性化命名， 可以避免通用指令 "你好手机"存在的被其他用户无意开启的 隐患）， 对语音段 "小 A等一下"、 "我通知小 B晚上一起吃饭" 的判断结果为否， 返 回步骤 S402, 对语音段 "你好小 5 " 的判断结果为是， 执行步骤 S404;
S404: 终端控制装置对具备开启属性的语音段进行判断， 判断是否合法， 是， 则 执行步骤 S405 , 否， 则返回步骤 S402; 判断语音段 "你好小 5 "的声纹与预设合法用户 （手机持有者） 的声纹是否一致， 若一致， 则语音段合法， 否则， 不合法；
S405 : 终端控制装置发送开启信号， 并开始后续语音段的存储； 向终端主语音系统发送开启信号 （该开启信号同时用于唤醒主操作系统 IOS及语 音系统 siri )， 并开始 "请电话小 B， …… " 的存储；
S406: 终端控制装置接收到开启成功信号， 检测是否正在进行某语音段的存储， 若是， 则执行步骤 S407, 若否， 则执行步骤 S408; 若在接收到 siri语音系统成功开启后所发送的开启成功信号时， 正在存储"晚上 6 点"这一语音段， 则执行步骤 S407, 若否， 则执行歩骤 S408; S407: 终端控制装置继续该语音段的存储； 并在该语音段的存储完成后， 执行步 骤 S408; 承接步骤 S406的假设， 在 "晚上 6点"存储完毕后 （语音中断大于 0.3秒）， 执 行步骤 S408; S408: 终端控制装置终止后续语音段的存储， 后续语音段交由主语音系统接收处 理， 并将存储的语音段传输出至主语音系统； 承接步骤 S407的假设， 后续语音段 "去某某餐厅吃饭" 由主语音系统 siri接收 (也可以由主操作系统 IOS接收后传输至 siri处理）； 同时将语音段 "请电话小 B "及 "晚上 6点"传输至终端主语音系统； S409: 主语音系统以各语音段的监听时间先后依次处理语音段， 并向用户展示； 主语音系统接收到的语音段以时间顺序依次为 "请电话小 B"、 "晚上 6点"及 "去 某某餐厅吃饭"， 主语音系统在网络侧的配合下依次处理向展示： 针对 "请电话小 B "， 终端主语音系统调用通讯录， 并向小 B发起呼叫； 针对 "晚上 6点"， 终端主语音系统建立日程安排 "晚上 6点有活动安排"； 针对"去某某餐厅吃饭"， 终端主语音系统配合网络侧在某某餐厅订餐， 查询交通 路线， 并向用户展示订餐结果及路线。 本实施例实现了主语音系统的自动开启、 语音段的连续存储等功能， 并且仅示例 性的给出了本发明的一个应用场景，在这个应用场景中执行步骤 S406的时机是主语音 系统成功开启后，在其他应用场景内，执行步骤 S406的时机是主操作系统成功唤醒后， 因为主操作系统的唤醒是在主语音系统开启之前的，这样，在主操作系统成功唤醒后， 由主操作系统来进行后续语音段的存储， 效果更好 （主操作系统接收存储语音段可以 降低对 DSP芯片的存储要求）。 综上可知， 通过本发明的实施， 至少存在以下有益效果： 在终端主语音系统处于关闭状态时， 持续监听包括至少一个语音段的用户语音， 针对用户语音中的每个语音段依次进行检测， 当检测到具备开启属性的语音段时， 向 终端主语音系统发送开启信号， 并同时开始存储在具备开启属性的语音段之后监听到 的语音段， 在终端主语音系统唤醒后， 将存储的语音段传输出去， 也即在检测到终端 主语音系统关闭后， 检测语音来判断是否需要开启终端主语音系统， 若需要， 则发送 开启信号， 实现了语音开启终端主语音系统， 同时在发送开启信号后， 并行存储后续 语音段， 用户不需要等待终端主语音系统开启， 即可连续说出语音指令， 也无需担心 语音指令终端主语音系统无法获得，解决了现有技术中语音控制技术操作繁杂的问题， 增强了用户的使用体验。 以上仅是本发明的具体实施方式而已， 并非对本发明做任何形式上的限制， 凡是 依据本发明的技术实质对以上实施方式所做的任意简单修改、等同变化、结合或修饰， 均仍属于本发明技术方案的保护范围。 工业实用性 本发明提供的上述技术方案， 可以应用于终端控制过程中， 采用在终端主语音系 统处于关闭状态时， 持续监听包括至少一个语音段的用户语音， 针对用户语音中的每 个语音段依次进行检测， 当检测到具备开启属性的语音段时， 向终端主语音系统发送 开启信号， 并同时开始存储在具备开启属性的语音段之后监听到的语音段， 在终端主 语音系统唤醒后， 将存储的语音段传输至终端主语音系统的技术方案， 用户不需要等 待终端主语音系统开启， 即可连续说出语音指令， 也无需担心语音指令终端主语音系 统无法获得， 解决了现有技术中语音控制技术操作繁杂的问题， 增强了用户的使用体 验。

Claims

权 利 要 求 书
1. 一种终端控制方法， 包括： 在检测到终端主语音系统关闭后， 监听包括至少一个语音段的用户语音； 判断是否出现具备开启属性的语音段；
若出现， 则发送用于开启所述终端主语音系统的开启信号， 并存储在所述 具备开启属性的语音段之后监听到的语音段；
在检测到所述终端主语音系统成功开启后， 将存储的语音段传输出去。
2. 如权利要求 1所述的终端控制方法， 其中， 所述判断是否出现具备开启属性的 语音段的步骤包括： 依次获取所述用户语音中各语音段的内容特征参数， 分别 与预设开启参数匹配，所述内容特征参数用于识别语音段的内容；若一语音段的 内容特征参数与所述预设开启参数匹配通过， 则该语音段具备开启属性。
3. 如权利要求 2所述的终端控制方法， 其中， 在发送所述开启信号之前还包括： 获取所述用户语音中各语音段的身份识别参数， 并分别与预设合法用户身份识 别参数比对,所述身份识别参数用于识别语音段的说话人身份；仅在出现内容特 征参数匹配通过、 身份识别参数比对合法的语音段时， 发送所述开启信号， 并 存储在所述具备开启属性的语音段之后监听到的语音段。
4. 如权利要求 1所述的终端控制方法， 其中， 在存储在所述具备开启属性的语音 段之后监听到的语音段之前， 还包括： 存储所述具备开启属性的语音段。
5. 如权利要求 4所述的终端控制方法， 其中， 在存储所述具备开启属性的语音段 之前， 还包括： 判断所述具备开启属性的语音段是否同时具备控制属性， 若是， 则存储所述具备开启属性的语音段。
6. 如权利要求 1至 5任一项所述的终端控制方法， 其中， 还包括： 在检测到所述终端主语音系统成功开启后， 判断是否正在存储语音段， 若 是， 则继续存储该语音段， 并在该语音段中断后， 终止后续语音段的存储， 若 否， 则终止后续语音段的存储；
在终止后续语音段的存储后，将后续语音段及存储的所有语音段传输出去。 如权利要求 6所述的终端控制方法， 其中， 若后续语音段和 /或存储的所有语音 段为多个语音段， 则所述终端主语音系统按照各语音段的监听时间先后， 依次 处理所述多个语音段或依次传输所述多个语音段至网络侧。 一种终端控制装置， 包括： 设置为在检测到终端主语音系统关闭后， 监听包括至少一个语音段的用户 语音的语音监听模块；
设置为判断是否出现具备开启属性的语音段的属性判断模块；
设置为若出现， 则发送用于开启所述终端主语音系统的开启信号， 并存储 在所述具备开启属性的语音段之后监听到的语音段的开启存储模块； 以及 设置为在检测到所述终端主语音系统成功开启后， 将存储的语音段传输出 去的数据传输模块。 如权利要求 8所述的终端控制装置， 其中， 所述数据传输模块包括： 设置为在检测到所述终端主语音系统成功开启后， 判断是否正在存储语音 段， 若是， 则继续存储该语音段， 并在该语音段中断后终止后续语音段的存储， 若否， 则终止后续语音段的存储的第一子模块的存储控制模块； 以及 设置为在终止后续语音段的存储后， 将后续语音段及存储的所有语音段传 输出去的传输控制模块。 一种语音控制装置， 包括监听模块、 检测模块、 控制模块及存储模块， 其中， 所述监听模块设置为在检测到终端主语音系统关闭后， 监听包括至少一个 语音段的用户语音， 并传输至所述检测模块；
所述检测模块设置为判断所述监听模块监听到的用户语音中是否出现具备 开启属性的语音段， 并在出现所述具备开启属性的语音段时， 触发所述控制模 块；
所述控制模块设置为在被所述检测模块触发时发送用于开启所述终端主语 音系统的开启信号， 并控制所述存储模块存储所述监听模块在所述具备开启属 性的语音段之后监听到的语音段， 还设置为在检测到所述终端主语音系统成功 开启后， 控制所述存储模块将存储的语音段传输出去； 所述存储模块设置为在所述控制模块的控制下存储或传输所述语音段。
11. 如权利要求 10所述的语音控制装置， 其中， 所述检测模块具体设置为： 依次获 取监听到的用户语音中各语音段的内容特征参数， 分别与预设开启参数匹配, 所述内容特征参数用于识别语音段的内容； 若一语音段的内容特征参数与所述 预设开启参数匹配通过， 则该语音段具备开启属性。
12. 如权利要求 11所述的语音控制装置，其中，所述检测模块还设置为在触发所述 控制模块之前， 获取各语音段的身份识别参数， 并分别与预设合法用户身份识 别参数比对,所述身份识别参数用于识别语音段的说话人身份；仅在出现内容特 征参数匹配通过、 身份识别参数比对合法的语音段时， 触发所述控制模块。
13. 如权利要求 10所述的语音控制装置，其中，所述控制模块还设置为在控制所述 存储模块存储所述监听模块在所述具备开启属性的语音段之后监听到的语音段 之前， 存储所述具备开启属性的语音段。
14. 如权利要求 13所述的语音控制装置，其中，所述控制模块还设置为在控制所述 存储模块存储所述具备开启属性的语音段之前， 判断所述具备开启属性的语音 段是否同时具备控制属性， 若是， 则控制所述存储模块存储所述具备开启属性 的语音段。
15. 如权利要求 10至 14任一项所述的语音控制装置， 其中， 所述控制模块还设置 为在检测到所述终端主语音系统成功开启后， 判断所述存储模块是否正在存储 任一语音段， 若是， 则控制所述存储模块继续存储该语音段， 并在该语音段中 断后终止后续语音段的存储， 若否， 则控制所述存储模块终止后续语音段的存 储； 在终止后续语音段的存储后， 将后续语音段传输至终端主语音系统， 并控 制所述存储模块将存储的所有语音段传输出去。
16. 一种终端， 包括终端主语音系统及如权利要求 10至 15任一项所述的语音控制 装置； 所述语音控制装置设置为在终端主语音系统休眠时， 监听包括至少一个 语音段的用户语音， 判断是否出现具备开启属性的语音段， 在出现所述具备开 启属性的语音段时， 开启所述终端主语音系统， 并存储在所述具备开启属性的 语音段之后监听到的语音段， 在所述终端主语音系统开启后， 将存储的语音段 交由所述终端主语音系统。
17. 如权利要求 16所述的终端， 其中， 若后续语音段和 /或存储的所有语音段为多 个语音段， 则所述终端主语音系统用于按照各语音段的监听时间先后， 依次处 理所述多个语音段或依次传输所述多个语音段至网络侧。
PCT/CN2014/085066 2014-06-11 2014-08-22 一种终端控制方法、装置、语音控制装置及终端 WO2015188459A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP14894647.8A EP3157003B1 (en) 2014-06-11 2014-08-22 Terminal control method and device, voice control device and terminal
US15/317,760 US20170110131A1 (en) 2014-06-11 2014-08-22 Terminal control method and device, voice control device and terminal

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410256788.2A CN105280180A (zh) 2014-06-11 2014-06-11 一种终端控制方法、装置、语音控制装置及终端
CN201410256788.2 2014-06-11

Publications (1)

Publication Number Publication Date
WO2015188459A1 true WO2015188459A1 (zh) 2015-12-17

Family

ID=54832782

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/085066 WO2015188459A1 (zh) 2014-06-11 2014-08-22 一种终端控制方法、装置、语音控制装置及终端

Country Status (4)

Country Link
US (1) US20170110131A1 (zh)
EP (1) EP3157003B1 (zh)
CN (1) CN105280180A (zh)
WO (1) WO2015188459A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106898352A (zh) * 2017-02-27 2017-06-27 联想(北京)有限公司 语音控制方法及电子设备

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107369445A (zh) * 2016-05-11 2017-11-21 上海禹昌信息科技有限公司 同时支持语音唤醒以及语音控制智能终端的方法
CN107450879A (zh) * 2016-05-30 2017-12-08 中兴通讯股份有限公司 终端操作方法及装置
CN106272481A (zh) * 2016-08-15 2017-01-04 北京光年无限科技有限公司 一种机器人服务的唤醒方法及装置
CN106847285B (zh) * 2017-03-31 2020-05-05 上海思依暄机器人科技股份有限公司 一种机器人及其语音识别方法
US10803872B2 (en) * 2017-08-02 2020-10-13 Panasonic Intellectual Property Management Co., Ltd. Information processing apparatus for transmitting speech signals selectively to a plurality of speech recognition servers, speech recognition system including the information processing apparatus, and information processing method
CN107861618A (zh) * 2017-11-10 2018-03-30 联想(北京)有限公司 电子设备控制方法和控制系统
CN108269570B (zh) * 2018-01-17 2020-09-11 深圳聚点互动科技有限公司 语音控制背景音乐主机的方法、装置、设备及存储介质
CN110689884A (zh) * 2019-09-09 2020-01-14 苏州臻迪智能科技有限公司 智能设备控制方法及装置
CN112744474A (zh) * 2019-10-29 2021-05-04 上海浦东临港智慧城市发展中心 一种用于公共场所的智能语音分类垃圾系统
CN111124512B (zh) * 2019-12-10 2020-12-08 珠海格力电器股份有限公司 智能设备的唤醒方法、装置、设备和介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202584717U (zh) * 2012-03-27 2012-12-05 北京腾瑞万里科技有限公司 语音识别装置及语音控制系统
CN103187051A (zh) * 2011-12-28 2013-07-03 上海博泰悦臻电子设备制造有限公司 车载互动装置
WO2013125916A1 (en) * 2012-02-24 2013-08-29 Samsung Electronics Co., Ltd. Method and apparatus for controlling lock/unlock state of terminal through voice recognition
CN103543814A (zh) * 2012-07-16 2014-01-29 瑞昱半导体股份有限公司 信号处理装置以及信号处理方法
CN103760969A (zh) * 2013-12-12 2014-04-30 宇龙计算机通信科技(深圳)有限公司 移动终端及语音控制应用程序的方法

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7054819B1 (en) * 2000-02-11 2006-05-30 Microsoft Corporation Voice print access to computer resources
US8060366B1 (en) * 2007-07-17 2011-11-15 West Corporation System, method, and computer-readable medium for verbal control of a conference call
CN102546953A (zh) * 2012-02-07 2012-07-04 深圳市金立通信设备有限公司 一种全语音操控移动终端的系统及方法
CN103327158A (zh) * 2012-03-19 2013-09-25 上海博路信息技术有限公司 一种语音识别的锁定和开锁方法
KR20130133629A (ko) * 2012-05-29 2013-12-09 삼성전자주식회사 전자장치에서 음성명령을 실행시키기 위한 장치 및 방법
KR102117841B1 (ko) * 2012-10-30 2020-06-02 삼성전자주식회사 전자장치 및 그 제어방법
CN103021411A (zh) * 2012-11-27 2013-04-03 威盛电子股份有限公司 语音控制装置和语音控制方法
US10395651B2 (en) * 2013-02-28 2019-08-27 Sony Corporation Device and method for activating with voice input
US9361885B2 (en) * 2013-03-12 2016-06-07 Nuance Communications, Inc. Methods and apparatus for detecting a voice command
US9343068B2 (en) * 2013-09-16 2016-05-17 Qualcomm Incorporated Method and apparatus for controlling access to applications having different security levels
CN103531201B (zh) * 2013-09-29 2016-04-06 上海云视科技股份有限公司 终端语音命名唤醒方法和系统
CN103595869A (zh) * 2013-11-15 2014-02-19 华为终端有限公司 一种终端语音控制方法、装置及终端
US9646607B2 (en) * 2014-03-10 2017-05-09 Dell Products, L.P. Managing wake-on-voice buffer quality based on system boot profiling

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103187051A (zh) * 2011-12-28 2013-07-03 上海博泰悦臻电子设备制造有限公司 车载互动装置
WO2013125916A1 (en) * 2012-02-24 2013-08-29 Samsung Electronics Co., Ltd. Method and apparatus for controlling lock/unlock state of terminal through voice recognition
CN202584717U (zh) * 2012-03-27 2012-12-05 北京腾瑞万里科技有限公司 语音识别装置及语音控制系统
CN103543814A (zh) * 2012-07-16 2014-01-29 瑞昱半导体股份有限公司 信号处理装置以及信号处理方法
CN103760969A (zh) * 2013-12-12 2014-04-30 宇龙计算机通信科技(深圳)有限公司 移动终端及语音控制应用程序的方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3157003A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106898352A (zh) * 2017-02-27 2017-06-27 联想(北京)有限公司 语音控制方法及电子设备

Also Published As

Publication number Publication date
EP3157003A4 (en) 2017-06-07
US20170110131A1 (en) 2017-04-20
CN105280180A (zh) 2016-01-27
EP3157003B1 (en) 2018-10-31
EP3157003A1 (en) 2017-04-19

Similar Documents

Publication Publication Date Title
WO2015188459A1 (zh) 一种终端控制方法、装置、语音控制装置及终端
AU2019246868B2 (en) Method and system for voice activation
JP6893951B2 (ja) 複数のデバイス上でのホットワード検出
TWI489372B (zh) 語音操控方法與行動終端裝置
CN110459221B (zh) 多设备协同语音交互的方法和装置
TWI535258B (zh) 語音接聽方法與行動終端裝置
WO2016161641A1 (zh) 语音识别方法、语音唤醒装置、语音识别装置及终端
CN107277272A (zh) 一种基于软件app的蓝牙设备语音交互方法及系统
EP3101533A1 (en) Mechanism for retrieval of previously captured audio
WO2016078214A1 (zh) 终端处理方法、装置及计算机存储介质
US20130085755A1 (en) Systems And Methods For Continual Speech Recognition And Detection In Mobile Computing Devices
CN103885350A (zh) 一种语音控制家庭电器的方法和装置
WO2015070644A1 (zh) 一种终端语音控制方法、装置及终端
WO2015027789A1 (zh) 语言控制方法、装置及终端
CN109410951A (zh) 基于Alexa语音控制的音响控制方法、系统及音响设备
US11178280B2 (en) Input during conversational session
CN107731231A (zh) 一种支持多云端语音服务的方法及一种存储设备
WO2016082344A1 (zh) 一种语音控制的方法、装置及存储介质
CN111862965A (zh) 唤醒处理方法、装置、智能音箱及电子设备
CN109903761A (zh) 语音交互方法、装置和存储介质
CN111524512A (zh) 低延时开启one-shot语音对话的方法、外围设备及低延时响应的语音交互装置
WO2019071723A1 (zh) 语音翻译方法、装置和翻译机
CN108093350B (zh) 麦克风的控制方法和麦克风
CN108847232A (zh) 一种处理方法及电子设备
CN109361820B (zh) 电子设备控制方法、系统及电子设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14894647

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 15317760

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2014894647

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2014894647

Country of ref document: EP