CN103092330B - Electronic device and method for speech recognition - Google Patents

Electronic device and method for speech recognition Download PDF

Info

Publication number
CN103092330B
CN103092330B CN 201110343163 CN201110343163A CN103092330B CN 103092330 B CN103092330 B CN 103092330B CN 201110343163 CN201110343163 CN 201110343163 CN 201110343163 A CN201110343163 A CN 201110343163A CN 103092330 B CN103092330 B CN 103092330B
Authority
CN
Grant status
Grant
Patent type
Application number
CN 201110343163
Other languages
Chinese (zh)
Other versions
CN103092330A (en )
Inventor
钟兴龙
柯杰斌
Original Assignee
宏碁股份有限公司
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
Grant date

Links

Abstract

一种语音辨识方法,应用于一电子装置。 A speech recognition method, applied to an electronic device. 该方法包括以下步骤:取得声音信号;转换声音信号为数字信号;自数字信号中取得多个数值,其中各数值代表对应的声波振幅值;依序比对各数值相对于至少一相邻数值的大小,以取得一组比对结果值;判断该组比对结果值是否对应至少一操作指令的其中之一;若是,执行对应的操作指令。 The method comprises the steps of: obtaining an audio signal; converts the sound signal into a digital signal; obtaining a plurality of values ​​from the digital signal, wherein each value represents the value corresponding to the amplitude of the acoustic wave; sequentially comparing each value with respect to at least one adjacent values size, to obtain a set of comparison result values; determines the set value comparison result whether an operation instruction corresponding to at least one of wherein; if so, performing a corresponding operation instruction. 通过本发明的技术方案,使用者所发出的声音信息经过本发明的信号转换及比对分析后,能够准确地判断此声音信息是否对应现有的操作指令,以确保能执行正确的操作指令,并且于使用上更加人性化。 By sound information aspect of the present invention, the user issued through the signal conversion of the present invention after analysis and comparison, can be accurately determined whether the sound information corresponding to this current operation instruction, can be performed to ensure proper operation instruction, and to use more humane.

Description

电子装置及其语音辨识方法 Electronic device and method for speech recognition

技术领域 FIELD

[0001] 本发明关于一种语音辨识方法,特别涉及一种应用于电子装置以便执行语音控制的语音辨识方法;此外本发明也包括应用该方法的电子装置。 [0001] The present invention relates to a speech recognition method, applied particularly to an electronic device to perform speech recognition of the voice control method; Furthermore, the present invention also includes an electronic device applying the method.

背景技术 Background technique

[0002] 随着语音辨识技术的日益精进,其应用的范围也越来越广泛。 [0002] With increasingly sophisticated speech recognition technology, the scope of its application has become increasingly widespread. 以电子装置为例,当结合语音辨识技术后,可提供使用者通过所发出的声音指令,来针对电子装置进行操控的功能;如此不仅简化了电子装置的操作,免除以往使用者必须通过各类遥控器输入指令,或是直接以手动触控电子装置的对应功能键所造成的不便,并且也提高了使用者使用上的便利性。 Electronic devices, for example, when combined with voice recognition technology to provide the user to manipulate the functions for electronic devices through voice commands issued; so not only simplifies the operation of the electronic device, the user must past through various exemptions remote control input command, directly or inconvenient manual function key corresponding to the electronic device caused by the touch, and also improves the convenience of users.

[0003]目前普遍采用的语音辨识技术,大多是直接将声音信号转换为电信号,并通过模拟数字转换器将声音信号波形数值化,以利于进行声音辨识及判断。 [0003] The speech recognition technology is currently widely used, most of the direct sound signal into an electrical signal by an analog-digital converter digitizing the sound signal waveform, performs a voice recognition and to facilitate the determination. 然而由于不同的译码器所产生的声音波形不尽相同,如何能针对现有波形进行有效分析,以减少声音辨识时的误判并提高辨识度,实为一值得研究的课题。 However, due to the different decoders generated sound waveform varies, how effectively the existing waveform analysis, in order to reduce false positives when the voice recognition and to improve recognition, a fact worth studying.

发明内容 SUMMARY

[0004] 本发明的主要目的的提供一种应用声音飞数值化转换以判断语音指令的语音辨识方法,以解决现有技术采用的语音辨识技术声音辨识时的误判较高的问题。 [0004] The main object of the present invention to provide a digitized voice applications fly conversion to a speech recognition method of determining voice commands, to solve a problem of a high false prior art speech recognition technology using voice recognition.

[0005] 为达到上述的目的,本发明采用以下技术方案: [0005] To achieve the above object, the present invention employs the following technical solution:

[0006] 一种语音辨识方法,应用于一电子装置,该方法包括以下步骤: [0006] A speech recognition method applied to an electronic device, the method comprising the steps of:

[0007] 取得一声音信号; [0007] obtaining an audio signal;

[0008] 转换该声音信号为一数字信号; [0008] converting the sound signal into a digital signal;

[0009]自该数字信号中取得多个数值,其中各该数值代表对应的一声波振幅值; [0009] obtaining a plurality of values ​​from the digital signal, wherein the acoustic wave amplitude value corresponding to each of the value represents;

[0010] 依序比对该多个数值中各该数值相对于至少一相邻数值的大小,以取得一组比对结果值; [0010] sequentially comparing the size of each of the plurality of numerical values ​​with respect to at least one adjacent values ​​to obtain a set of comparison result values;

[0011] 判断该组比对结果值是否对应至少一操作指令的其中之一;以及 [0011] Analyzing the results of the comparison value is set corresponding to one of at least one operating instruction therein; and

[0012] 当判断该组比对结果值对应该至少一操作指令的其中之一时,执行对应的该操作指令。 [0012] When the operation instruction set determines the ratio of the resulting value should be at least one of an operating instruction which executes a corresponding.

[0013] 较佳地,该声音信号通过傅立叶转换而形成该数字信号。 [0013] Preferably, the audio signal is formed by Fourier transformation of the digital signal.

[0014] 较佳地,通过过滤掉该数字信号中低于一标准值的信号值,以取得该多个数值。 [0014] Preferably, by filtering out the signal value of the digital signal is lower than a standard value, to obtain the plurality of values.

[0015] 较佳地,还包括以下步骤:检测是否接收一启动信号,以判断是否开始取得该声音信号。 [0015] Preferably, further comprising the steps of: detecting whether a start signal is received, to determine whether to start acquisition of the audio signal.

[0016] 较佳地,于判断该组比对结果值是否对应至少一操作指令的其中之一的步骤中,通过比对该至少一操作指令所对应的至少一组参考值与该组比对结果值是否相符或数值差异量在一定范围内,以判断该组比对结果值是否对应该至少一操作指令的其中之一。 [0016] Preferably, the set is determined in the comparison result values ​​corresponds to at least one step in which an operation instruction, at least one set of reference values ​​by comparing the at least one operating instruction corresponding to the set of alignment the results are consistent with the value of difference in the amount or value within a certain range, to determine whether the set ratio should be at least one instruction of an operation in which the result values.

[0017] 较佳地,将各该数值均与同一侧的一单一相邻数值加以比对,依据各该数值相对于该单一相邻数值的大小状态,以取得对应的该组比对结果值。 [0017] Preferably, each of the average value of the ratio to be adjacent to a single side of the same value, based on each of the adjacent single value with respect to the size of the state values ​​to obtain the set of values ​​corresponding to the comparison result .

[0018] 较佳地,将各该数值同时与位于该数值两侧的该两相邻数值加以比对,依据各该数值相对于该二相邻数值的大小状态,以取得对应的该组比对结果值。 [0018] Preferably, the ratio of each of the values ​​to be simultaneously on the two adjacent sides of the value of the value is in, based on the value relative to the size of each state of the two adjacent values ​​to obtain a corresponding set of the ratio the result value.

[0019] 较佳地,还包括以下步骤:当判断该组比对结果值未对应该至少一操作指令的其中之一时,检测是否接收一设定信号;若是,设定该声音信号以对应一新操作指令。 [0019] Preferably, further comprising the step of: determining when one of the comparison result the set value is to be at least one operating instructions, detecting whether a setting signal; if so, sets the audio signal to correspond to a The new operating instructions.

[0020] 本发明还提供一种电子装置,包括: [0020] The present invention also provides an electronic device, comprising:

[0021] —声音接收模块,用以接收一声音信号;以及 [0021] - a sound receiving module for receiving an audio signal;

[0022] —处理模块,与该声音接收模块电性连接,该处理模块包括: [0022] - a processing module, electrically connected to the sound receiving module, the processing module comprising:

[0023] —转换单元,用以转换该声音信号为一数字信号,并自该数字信号中取得多个数值,其中各该数值代表对应的一声波振幅值;以及 [0023] - converting means for converting the sound signal into a digital signal, and obtaining a plurality of values ​​from the digital signal, wherein each of the sound wave amplitude value corresponding to the representative value; and

[0024] —比对编码单元,用以依序比对该多个数值中各该数值相对于至少一相邻数值的大小,以取得一组比对结果值; [0024] - comparing the coding unit for sequentially comparing the size of each of the plurality of numerical values ​​with respect to at least one adjacent values ​​to obtain a set of comparison result values;

[0025] 通过该处理模块判断该组比对结果值是否对应至少一操作指令的其中之一,以决定是否执行对应的该操作指令。 [0025] wherein one determines the set value comparison result whether an operation instruction corresponding to the at least by the processing module to determine whether the operation corresponding to the command execution.

[0026] 较佳地,该处理模块通过傅立叶转换将该声音信号转换成该数字信号。 [0026] Preferably, the processing module converts the digital signal into a sound signal into the Fourier.

[0027] 较佳地,该转换单元过滤掉该数字信号中低于一标准值的信号值,以取得该多个数值。 [0027] Preferably, the conversion unit filtered signal value is lower than a standard value of the digital signal, to obtain the plurality of values.

[0028] 较佳地,还包括一启动模块,用以产生一启动信号至该处理模块,以通知该声音接收模块开始取得该声音信号。 [0028] Preferably, further comprising a starting module for generating a start signal to the processing module, to inform the sound receiving module to obtain the audio signal starts.

[0029] 较佳地,还包括一存储模块,用以存储该至少一操作指令所对应的至少一组参考值,其中当该处理模块比对该至少一组参考值的其中之一与该组比对结果值相符或数值差异量在一定范围内时,则判断该组比对结果值对应该至少一操作指令的其中之一。 [0029] Preferably, further comprising a storage module for storing at least one set of at least one reference value corresponding to the operation instruction, wherein when the processing module than the one of the at least one set of reference values, wherein the set of values ​​or the comparison result matches the numerical difference amount within a certain range, it is determined that the set value of the ratio should be at least one of a result of which the operational instructions.

[0030] 较佳地,将各该数值均与同一侧的一单一该相邻数值加以比对,依据各该数值相对于该单一相邻数值的大小状态,以取得对应的该组比对结果值。 [0030] Preferably, each of the values ​​are the same with a single value to be adjacent to the side of the comparison, according to each of the adjacent single value with respect to the size of the state values, the set to obtain a corresponding comparison result value.

[0031] 较佳地,将各该数值同时与位于该数值两侧的该二相邻数值加以比对,依据各该数值相对于该二相邻数值的大小状态,以取得对应的该组比对结果值。 [0031] Preferably, the value of each of the two simultaneously located adjacent the sides of the value of the value to be aligned, according to each of the state value relative to the size of the two adjacent values ​​to obtain a corresponding set of the ratio the result value.

[0032] 进一步地,当该处理模块判断该组比对结果值未对应该至少一操作指令的其中之一时,该处理模块检测是否接收一设定信号,以设定该声音信号以对应一新操作指令。 [0032] Further, when the processing module determines the set to be not more than one of the at least one operational instructions wherein the result value, the processing module detecting whether a setting signal to set the sound signal to correspond to a new operation instruction.

[0033] 借此设计,使用者所发出的声音信息经过本发明的信号转换及比对分析后,能够准确地判断此声音信息是否对应现有的操作指令,以确保能执行正确的操作指令,并且于使用上更加人性化。 [0033] The design whereby the sound information, the user issued through the signal conversion of the present invention after analysis and comparison, can be accurately determined whether the sound information corresponding to this current operation instruction, can be performed to ensure proper operation instruction, and to use more humane.

附图说明 BRIEF DESCRIPTION

[0034] 图1是本发明的电子装置的示意图; [0034] FIG. 1 is a schematic diagram of an electronic device according to the present invention;

[0035]图2(a)是本发明的电子装置所接收的声音信号的波形图; [0035] FIG. 2 (a) is a waveform of a voice signal of an electronic device of the present invention is received;

[0036]图2(b)本发明的电子装置所接收的声音信号经转换后所形成的波形图; Waveform (b) of the electronic device according to the present invention, the received sound signal after conversion of the formed [0036] FIG 2;

[0037] 图2(c)是本发明的电子装置将比对结果值与参考值相比对的示意图; [0037] FIG. 2 (c) is an electronic schematic diagram of apparatus of the invention than the result of comparison of the values ​​with a reference value;

[0038]图3是本发明的语音辨识方法的第一实施例的流程图; [0038] FIG. 3 is a flowchart of a first embodiment of the speech recognition method of the present invention;

[0039]图4是本发明的语音辨识方法的第二实施例的流程图; [0039] FIG. 4 is a flowchart of a second embodiment of the speech recognition method of the present invention;

具体实施方式 detailed description

[0040] 为能让审查员能更了解本发明的技术内容,特举出较佳实施例说明如下。 [0040] The examiner can make a better understanding of the technical content of the present invention, particularly preferred embodiments include embodiments described below.

[0041] 请先参考图1是本发明的电子装置I的示意图。 [0041] Please refer to FIG. 1 is a schematic diagram of the electronic device I of the present invention. 在本发明的一实施例中,电子装置I是一遥控器以针对家用电器进行远程控制,而依据应用类型不同,电子装置I也可为一家用电器(例如电视或影音播放机等)、一可携式电子产品(例如计算机或手机等)或其它可应用声控方式操作的电子装置,但不以本实施例为限。 In an embodiment of the present invention, the electronic device I is a remote controller for remote control for household appliances, and depending on the type of application, the electronic device I may also be a household appliance (e.g. a TV or video player, etc.), a the electronic device portable electronic products (e.g. computer or mobile phone, etc.) or other applications may be a voice mode of operation, but is not limited to the present embodiment.

[0042] 如图1所示,本发明的电子装置I包括声音接收模块10、处理模块20及存储模块30。 [0042] As shown in FIG. 1, the electronic device according to the present invention I include a sound receiving module 10, a processing module 20 and memory module 30. 声音接收模块10用以接收使用者所发出的声音信号,在本发明的一实施例中,声音接收模块10为一声音转化器或一麦克风,但也可采用其它具有类似收音功能的装置。 Sound receiving sound signals emitted for receiving a user module 10, in one embodiment of the present invention, the receiving module 10 is a sound converter or a sound microphone, but other similar devices having a radio function.

[0043] 处理模块20与声音接收模块10电性连接,处理模块20包括转换单元22及比对编码单元24。 [0043] The processing module 20 and the sound receiving module 10 is electrically connected to the processing module 20 includes a conversion unit 22 and coding unit 24 ratio. 转换单元22用以转换已接收的声音信号为数字信号,其中转换单元22通过傅立叶转换方式将声音信号转换成对应的数字信号,例如数字模拟转换器。 Audio signal converting unit 22 to convert the received digital signal, wherein the conversion unit 22 by way of Fourier transformation to convert the audio signal into a corresponding digital signal, such as a digital-analog converter.

[0044] 请一并参考图1及图2(a)、(b)及(C)。 [0044] Please refer to FIG. 1 and FIG. 2 (a), (b), and (C). 图2 (a)是本发明的电子装置I所接收的声音信号的波形图,图2(b)是本发明的电子装置I所接收的声音信号经转换后所形成的波形图,图2(c是本发明的电子装置I将比对结果值与参考值相比对的示意图。如图2(a)所示,本发明的电子装置I的声音接收模块10在接收到使用者所发出的声音信号后,即可取得如图所示声波的时间-振幅波形图,由于此声音信号过于杂乱而难以辨识,因此需要经过信号的转换以便提高辨识效果。如图2(b)所示,经傅立叶转换后所产生声波的频率-振幅波形图,将使得数字信号会呈现多个高低不一的振幅峰值,本发明的技术即是针对此振幅峰值的数值处理以达到语音辨识的效果。 FIG 2 (a) is a waveform of a voice signal I received an electronic device according to the present invention, FIG. 2 (b) is a waveform diagram of the electronic device according to the present invention I received sound after the converted signal is formed, FIG. 2 ( c is a schematic view of an electronic device according to the present invention compared with the results than I value and the reference value is shown in Figure 2 (a), the electronic device according to the present invention, the I sound receiving module 10 receives the issued user after the speech signal, the acoustic wave shown in FIG time to get - the amplitude of the waveform of FIG. Since this sound signal is too cluttered and difficult to identify, the converted signal is required in order to improve recognition results shown in FIG. 2 (b), by. frequency acoustic waves generated by the Fourier Transform - amplitude waveform diagram, the digital signal so that a plurality of amplitude peaks will appear uneven, the techniques of the present invention is directed, i.e., the peak amplitude value of this process in order to achieve voice recognition.

[0045] 如图2(b)所示,转换单元22自经转换后的数字信号中取得多个数值,而在欲取得这些数值的过程中,为避免大量振幅峰值较低的数值影响判断的准确度,因此转换单元22先依已设定的一振幅数值做为标准值(如图中虚线所示),并针对数字信号中所取得的多个峰值加以过滤,删除掉低于前述标准值的峰值以避免噪声干扰,最后取得余留下来的多个数值;其中各数值代表对应的声波振幅值。 [0045] FIG. 2 (b), the process of obtaining a plurality of values ​​from a digitally converted signal, and in To obtain these values ​​conversion unit 22 in order to avoid large peak amplitude value to influence the judgment of the lower accuracy, thus converting unit 22 as the first reference value (broken line shown in FIG.) in accordance with an amplitude value has been set, and filtered digital signals for a plurality of peaks obtained, the standard value is lower than the deleted to avoid interference noise peaks, and finally left to obtain a plurality of values ​​I; wherein each numerical value represents the amplitude of the acoustic wave corresponding to. 在本发明的一实施例中,转换单元22自经转换后的数字信号中取得M个数值,且M为大于I的整数。 In an embodiment of the present invention, the conversion unit 22 obtains the M value from the digital signal after conversion, and M is an integer greater than I.

[0046] 比对编码单元24用以对多个数值中的各数值,依序比对各数值相对于至少一相邻数值的大小,以取得一组比对结果值。 [0046] The coding unit 24 for each value of the plurality of numerical values, each numerical value sequence alignment relative to the size of the at least one neighboring value in order to obtain the result of a set of values ​​than the ratio. 此处所述依序比对各数值相对于至少一相邻数值的设计,是指将多个数值中的各个数值按照前后顺序各别与其相邻的左右数值进行比对,例如将各数值依序均与同一侧(左侧或右侧)的单一相邻数值两两比对,或是依序将各数值同时与左右两侧的相邻数值一齐比对;每个数值经前述比对后可产生一个比对结果,而在多个数值均经过比对后,即取得由多个比对结果组成的一组比对结果值。 Here the design values ​​are sequentially compared to each of at least one adjacent values ​​relative to the respective value refers to the plurality of values ​​in order to compare before and after the respective adjacent thereto according to the value of about, for example, by numerical values each value after the comparison by; single sequence are adjacent to the same side (left or right) values ​​pairwise alignments, or simultaneously and sequentially to the numerical value adjacent to the left and right sides together than It may generate a comparison result, and after each comparison, i.e. to obtain a set of a plurality of the comparison result by the ratio of the composition to result in a plurality of numerical values. 在本发明的一实施例中,假设比对编码单元24针对M个数值中的第N个数值相对于其左侧相邻的第(N-1)个数值依序进行两两比对,以取得一组比对结果值,其中N为整数且2彡N彡M。 In an embodiment of the present invention, it is assumed that the coding unit 24 for comparing the M values ​​in the N-th value of the left relative to its adjacent first (N-1) th numerical sequence for pairwise alignments to obtaining a set of comparison result values, where N is an integer and N 2 Pie San M. 如图2 (b)所示,举例来说,若转换单元22取得5个数值(此时M等于5,且N为2至5之间的任意数),比对编码单元24会依序比对第2个数值对第I个数值的大小,接着比对第3个数值对第2个数值的大小...依此类推,最后比对第5个数值对第4个数值的大小,进而得到一组比对结果值。 As shown in FIG 2 (b) as shown, for example, if the conversion value 22 to obtain 5 units (in this case equal to 5 M, and N is any number between 2-5), than the coding unit 24 will be sequentially ratio Subsequently, the final size of the size of the second I-th numerical value than the value of the third and so on ... the size of the second value than the first 5 values ​​of the fourth value, and further to obtain a set of comparison result values.

[0047] 其中对应前述数值大小的比较结果,可定义其结果分别对应不同的比对结果值。 [0047] wherein the values ​​corresponding to the size of the comparison result, the result may be defined corresponding to different values ​​of the ratio results. 当比对编码单元24判断第N个数值大于第(N-1)个数值时,记录第N个数值为一第一比对结果值;当判断第N个数值等于第(N-1)个数值时,记录第N个数值为一第二比对结果值;当判断第N个数值小于第(N-1)个数值时,记录第N个数值为一第三比对结果值。 When the specific coding unit 24 determines the N value is greater than a first value of (N-1), a record of the number N is a first comparison result value; when it is judged equal to the value of the N (N-1) th when the value of the number N of recording a second value comparison result values; when the value is less than determination of the N-th (N-1) th value of the number N of recording a third value comparison result values. 延续前例,假设定义第一比对结果值为1,第二比对结果值为2,第三比对结果值为O ;在取得5个数值的状况下,当第2个数值大于第I个数值时,记录第2个数值的比对结果值为I ;当第3个数值大于第2个数值时,记录第3个数值的比对结果值为I ;当第4个数值等于第3个数值时,记录第4个数值的比对结果值为O ;最后当第5个数值小于第4个数值时,记录第5个数值的比对结果值为2。 Continuation of the previous example, consider the definition of the first comparison result value is 1, the second comparison result is 2, and the third comparison result is O; under conditions to obtain five values, when the second value is greater than the I-th when the value recorded in the second comparison result value is I; when the third value is greater than the second value, the third value records comparison result is I; when the fourth value is equal to the third when the value is recorded more than four values ​​of the results is O; Finally, when the fifth value is less than the fourth value fifth value than the recording of the results is 2. 因此,比对编码单元24经数值依序两两比对后,会取得由前述四个比对结果所组成的一组比对结果值为” 1120”。 Thus, the ratio of the value 24 is sequentially pairwise alignments on the coding unit, will be acquired by the ratio of a set of four comparison result is composed of the results "1120." 通过所获得的此组比对结果值,即可作为后续指令比对的判断来源。 This group obtained comparison result values, to determine the source of a subsequent instruction alignment.

[0048] 在本发明的另一实施例中,若是在各数值同时与左右两侧的相邻数值一齐比对的设计下,假设比对编码单元24针对M个数值中的第N个数值相对于其左侧相邻的第(N-1)个数值及右侧相邻的第(N+1)个数值依序进行一齐比对,以取得一组比对结果值,其中N为整数且2彡N彡M-1。 [0048] In another embodiment of the present invention, if at the same time in each numerical value adjacent to the left and right sides than the design together, assuming the encoding unit 24 relative ratio of the M values ​​for the N-th value its left adjacent first (N-1) th value of the right adjacent first and (N + 1) th numerical sequence to be compared together to obtain a set of comparison result values, where N is an integer and N 2 San San M-1. 举例来说,若转换单元22取得如图2(b)所示的5个数值(此时M等于5,且N为2至4之间的任意数),比对编码单元24会依序比对第2个数值相对第I个数值及第3个数值的大小,接着比对第3个数值相对第2个数值及第4个数值的大小,最后比对第4个数值相对第3个数值及第5个数值的大小,进而得到一组比对结果值。 For example, if the conversion unit 225 to obtain numerical values ​​shown in FIG. 2 (b) (in this case equal to 5 M, and N is any number between 2 to 4), compared to the encoding unit 24 will be sequentially ratio , then the relative size of the second and fourth numerical values ​​of the I-th second and third numerical value than the third value relative to the second numerical value, the last four values ​​are compared to the first three values ​​of the relative 5 and the second size value, thereby to obtain a set of comparison result values.

[0049] 其中依据前述数值大小的比较结果,可分别定义出对应九种不同结果的比对结果值,其略分为三个数值相等(I种)、任两个数值相等且搭配另一数值的大小变化出种),三个数值均不相等(2种)等态样,也可使用数字I至9来表示各个态样所对应的比对结果值。 [0049] wherein based on the comparison result of numerical values, respectively, define values ​​corresponding to the comparison result of nine different results, which slightly into equal three values ​​(I species), and the two values ​​are equal to any other value with change the size of species), three values ​​are not equal (two kinds) and other aspects, may also be used to represent the digital I to 9 corresponding to the respective aspect ratio of the resulting value. 因此若依图2 (b)所取得的5个数值来判断,假设定义三个数值均不相等且第2个数值大于第I个数值但小于第3个数值的比对结果值为8,第3个数值等于第4个数值且大于第2个数值的比对结果值为2,而第4个数值等于第3个数值且大于第5个数值的比对结果值为5。 Thus numerical Zoe FIG. 25 (b) to determine the acquired, assume three values ​​are not equal and defined second value is greater than the I-th numerical value but less than the third comparison result is 8, 3 Numerical value is equal to the fourth value and greater than the second comparison result is 2, and the fourth value is equal to and greater than the third value fifth value than the value of 5 results. 因此,比对编码单元24经各数值与两侧相邻数值依序比对后,会取得由前述三个比对结果所组成的一组比对结果值为”825”,以此类推。 Thus, coding unit 24 by more than the value of the adjacent sides of each numerical sequence alignments, we will achieve a set of comparison result value "825" of the three comparison result composed, and so on. 前述各比对结果值可依不同的辨识设计而改变,本发明不以此为限。 Each value of the comparison result of the identification according to different design change, the present invention is not limited thereto.

[0050] 须注意的是,本发明的电子装置I的处理模块20可为一处理器或其与软件的结合,但处理模块20也可为单纯的硬件、软件、硬体或前述至少两者以上的组合,本发明不以此为限。 [0050] It should be noted that at least two of the electronic device I processing module 20 of the present invention may be a processor, or in combination with software, but the processing module 20 may also be a simple hardware, software, or the hardware combinations of the above, the present invention is not limited thereto.

[0051] 存储模块30用以存储前述该组比对结果值及至少一操作指令所对应的至少一组参考值。 [0051] The storage module 30 for storing the comparison result the set value and the at least one set of at least one reference value corresponding to the operation instruction. 其中使用者可预先设定并记录至少一声音指令以对应至少一操作指令,例如预录”关机”或”开机”的声音指令以对应本发明的电子装置I的关机或开机操作。 Wherein the user may set in advance and recording at least a voice command corresponding to at least an operation instruction to, for example, pre-recorded "off" or "on" corresponding to the voice instructions to the electronic device I of the present invention is shut down or start operation. 而这些声音指令一样可通过前述针对声音信号的处理方式,进行信号转换及比对分析,以取得其对应的至少一操作指令的至少一组参考值。 These may be the same as for the voice command processing mode of the sound signal, performing signal conversion and comparison analysis to obtain at least one operational instructions corresponding to at least one of the set of reference values. 例如经信号转换比对后,”关机”的操作指令对应于一组参考值”2101”,而”开机”的操作指令则对应于另一组参考值” 1121”。 After signal conversion, for example, alignment, "off" operation instruction corresponding to a set of reference values ​​"2101", and "power" is an operation instruction corresponding to another set of reference values ​​"1121." 这些参考值也会被存储于存储模块30中以供后续比对使用。 The reference value is also stored in the storage module 30 for subsequent use alignments.

[0052] 借此,处理模块20依据所取得的该组比对结果值判断其是否对应于前述任一组参考值,以确认该组比对结果值是否对应任一操作指令,进而执行所对应的操作指令。 [0052] Accordingly, the set of the processing module 20 based on the acquired comparison result determines whether or not the value of which corresponds to one of the preceding set of reference values, to confirm the set value comparison result corresponds to any one of the operation instructions, and further execute corresponding operation instruction. 其中当处理模块20比对任一组参考值与该组比对结果值相符或数值差异量在一定范围内时,则判断该组比对结果值对应该至少一操作指令的其中之一,进而执行对应的操作指令。 Wherein when the difference in the amount of the processing module 20 to match any of a set of reference values ​​with the set values ​​or the comparison result value is within a certain range, it is determined that the set value of the comparison result should be at least one of an operation instruction, wherein, further executes a corresponding operation instruction. 如图2 (c)所示,延续前例,假设当处理模块20判断任一组参考值与该组比对结果值数值完全相同或两者数值差异量为I个数值以内,当已取得的该组比对结果值为”1120”分别与代表关机的该组参考值”2101”及代表开机的该组参考值”1121”加以比对后,可得知该组比对结果值”1120”与代表开机的该组参考值”1121”较为相近且仅有I个数值的差异,因此处理模块20即可判断该组比对结果值对应开机的操作指令,借此依据该声音信息执行电子装置I的开机操作。 As shown in FIG 2 (c), the continuation of the previous example, assume that when any of the processing module 20 determines a set of reference values ​​with the set value comparison result values ​​are identical, or the amount of difference between the two values ​​is within the I value, when the acquired after the value is "1120", respectively, representing the set of reference values ​​off the set of reference values ​​"2101" and the power of the representative of "1121" to be compared with a comparison result set, the set of alignment that can result value "1120" and representative of the set reference power value "1121" and only the differences are more similar to the I value, the processing module 20 to the comparison result determines the set value corresponding to the start of the operation instruction, thereby performing the sound information based on the electronic device I boot operation. 须注意的是,前述设定的比对数值差异量也可依辨识设计需求而改变,本发明不以此为限。 It should be noted that the ratio of the amount of the difference value is set also follow identify design requirements vary, according to the present invention is not limited thereto. 在本发明的一实施例中,存储模块20可为内存或其它可供存储的装置。 In an embodiment of the present invention, the storage module 20 is a memory or other means may be available for storage.

[0053] 此外,当处理模块20依据前述判断机制判断该组比对结果值并未对应任一操作指令时,表示使用者所发出的声音信号与任一操作指令均不相符,此时处理模块20可通过其它器件或模块发出提示信息,例如预设的声响或闪光等,用以告知使用者。 [0053] Further, when the processing module 20 based on the judgment mechanism determines the set value comparison result does not correspond to any one time operation instruction, a signal representing sound emitted by the user with any operation instruction not match, then the processing module 20 may be sent by other devices or modules prompt information, or the like, for example, flash a predetermined sound to inform the user. 此时使用者可通过电子装置I的特定按键输入一设定信号,当处理模块20检测已接收此设定信号时,即可依据此设定信号将所接收的声音信号赋予新的对应设定,使其对应一新操作指令。 At this time, the user can input a setting signal of the electronic device through a specific key I, when the processing module 20 detects this setting signal is received, according to this setting signal to the sound signal corresponding to the received set to give new , it corresponds to a new operating instructions. 借此,当语音辨识效果不理想而无法分辨使用者时声音时,使用者能自行重新录制及设定目前所发出的声音信息及其对应的操作指令,提高使用上的便利性。 When this, when the voice recognition result is not satisfactory and can not distinguish the sound when the user, the user can re-record their own voice message and its corresponding set of operating instructions currently issued, improving the convenience of use.

[0054] 本发明的电子装置I还包括启动模块40,用以产生一启动信号至处理模块20,以通知声音接收模块10开始取得使用者所发出的声音信号。 [0054] The electronic device according to the present invention further comprises I startup module 40, for generating a start signal to the processing module 20 to notify the sound receiving module 10 starts to acquire sound signal emitted by a user. 在本发明的一实施例中,启动模块40可为一输入按键或以其它具类似功能的器件所取代,不以本实施例为限。 In an embodiment of the present invention, the boot module 40 may be a substituted or input keys with similar functions in other devices, not limited to the present embodiment.

[0055] 请参考图3是本发明的语音辨识方法的第一实施例的流程图。 [0055] Please refer to FIG. 3 is a flowchart of a first embodiment of the speech recognition method of the present invention. 须注意的是,以下虽以图1所示的电子装置I为例说明本发明的语音辨识方法,但本发明并不以适用于电子装置I为限,任何其它具类似架构的电子装置也可适用本发明的语音辨识方法。 It should be noted that, although the electronic device I shown in FIG. 1 as an example of the speech recognition method of the present invention, but the present invention is not to be suitable for electronic device I is limited, any other electronic device with a similar architecture may also be speech recognition method of the present invention is applied. 如图3所示,本发明的语音辨识方法包括步骤S301至步骤S306。 As shown in FIG. 3, the speech recognition method of the present invention comprises a step S301 to step S306. 以下将详细说明该方法的各个步骤。 The various steps of the method described in detail below.

[0056] 步骤S301:取得一声音信号。 [0056] Step S301: obtaining a sound signal. 电子装置I通过声音接收模块10来取得使用者所发出声音信号,以进行后续处理。 The electronic device I received by a sound module 10 to acquire the sound signal emitted by a user, for subsequent processing.

[0057] 步骤S302:转换声音信号为数字信号。 [0057] Step S302: the sound signal is converted into a digital signal. 通过处理模块20的转换单元22针对声音信号进行傅立叶转换,已取得对应的数字信号。 22 by the processing module 20 for converting a sound signal unit Fourier transform has been made a corresponding digital signal.

[0058] 步骤S303:自数字信号中取得多个数值。 [0058] Step S303: obtaining a plurality of values ​​from the digital signal. 转换单元22自转换后的数字信号中,先过滤掉低于一标准值的振幅峰值,以减少其所造成类似噪声的误判影响,接着自剩余振幅峰值中取得多个数值,其中各数值代表对应的声波振幅值。 Digital signal conversion from the conversion unit 22, to filter the peak amplitude is below a standard value, in order to reduce their impact noise-like false, then obtaining a plurality of amplitude values ​​from the remaining peaks, where each value represents corresponding to the acoustic wave amplitude value.

[0059] 步骤S304:依序比对各个数值相对于至少一相邻数值的大小,以取得一组比对结果值。 [0059] Step S304: sequentially comparing each value with respect to at least the size of a neighboring values, the result values ​​to obtain a set ratio. 处理模块20的比对编码单元24依据前述所取得的多个数值,针对各个数值与至少一相邻数值相互比对,并依比对结果不同而赋予不同的比对结果值,借此以取得一组比对结果值。 Processing module 20 than the coding unit 24 based on the plurality of acquired values, for each numerical value and at least one adjacent another alignment, and different depending on the comparison result of the result value to impart a different ratio, thereby to obtain a set of comparison result values.

[0060] 步骤S305:判断该组比对结果值是否对应至少一操作指令的其中之一。 [0060] Step S305: determining whether the comparison result value is set corresponding to one wherein the at least one operation instruction. 处理模块20依据所取得的该组比对结果值与至少一操作指令所对应的至少一组参考值相比对,判断任一组参考值与该组比对结果值是否相符或数值差异量在一定范围内。 The acquired group 20 according to the processing module and at least one operation instruction corresponding to the at least one set of reference values ​​of the comparison result values, determining any of a set of reference values ​​with the set value comparison result values ​​are consistent or the amount of difference in within a certain range.

[0061] 步骤S306:执行对应的操作指令。 [0061] Step S306: performing a corresponding operation instruction. 当处理模块20判断有任一组参考值与该组比对结果值是否相符或数值差异量在一定范围内时,即表示该组比对结果值所对应的声音信号代表该组参考值所对应的操作指令,因此处理模块20可执行对应的操作指令。 When the processing module 20 determines if any one set of reference values ​​are consistent with the set of comparison results or numerical difference amount within a certain range, i.e., indicates that the group than the sound signal result value corresponding to the representatives of the group reference value corresponding to operation instruction, so the processing module 20 may perform a corresponding operation instruction.

[0062] 借此设计,将使用者所发出的声音信息通过信号转换、过滤、比对分析等处理,来判别此声音信息是否与现存的操作指令相符,以决定是否执行对应的操作指令。 [0062] whereby the design, the sound emitted by the user information signal conversion, filtering, comparison analysis process, discriminates whether the sound information is consistent with existing instruction operation to determine whether the corresponding operation instruction execution.

[0063] 请参考图4是本发明的语音辨识方法的第二实施例的流程图。 [0063] Please refer to FIG. 4 is a flowchart of a second embodiment of the speech recognition method of the present invention. 如图4所示,其中步骤S401至步骤S406与前述步骤S301至步骤S306相同,而本实施例还包括步骤S407、S408及至步骤S409。 4, the same step S401 to step S406 where the aforementioned step S301 to step S306, the present embodiment further comprises the step of S407, S408 till step S409.

[0064] 于步骤S401之前还包括步骤S407:检测是否接收一启动信号。 [0064] The step S401 further comprising before the step S407: detecting whether a start signal. 当使用者欲使用电子装置I执行本发明的语音辨识方法时,先通过处理模块20检测是否接收使用者通过启动模块40所输入的启动信号,以决定是否启动声音撷取的相关功能。 When the user wants to use the electronic device while performing a voice recognition method I of the present invention, whether to receive the first user by the processing module 20 detects the starting signal input 40 starting module to determine whether to start capturing the voice-related functions. 当检测到此启动信号时,处理模块20即通知声音接收模块10开始进行声音撷取。 When this start signal is detected, i.e., the processing module 20 notification module 10 to start the sound receiving sound capturing.

[0065] 于步骤S405之后还包括步骤S408:检测是否接收一设定信号。 [0065] In step S405, after further comprising the step S408: detecting whether a setting signal. 当通过步骤S405判断该组比对结果值并未对应任一操作指令时,处理模块20检测是否接收到使用者所输入的设定信号,以判断是否针对该组比对结果值所对应的声音信号重新进行操作指令的定义。 When step S405 determines the set value comparison result does not correspond to any one of the operational instructions, the processing module 20 detects whether the received signal to a set input by the user to determine whether the results for the set of ratio values ​​corresponding to the sound operation instruction signal is re-defined.

[0066] 于步骤S408之后还包括步骤S409:设定声音信号以对应一新操作指令。 [0066] In step S408, after further comprising the step S409: the sound signal is set to correspond to a new operation command. 当处理模块20检测到使用者所输入的设定信号时,则依据此设定信号将声音信息重新定义以对应新的操作指令。 When the processing module 20 detects a setting signal input by the user, the sound signal based on this setting information corresponding to redefine a new operation command.

[0067] 通过本发明的设计,除了应用于遥控器或类似装置上,让使用者可通过语音输入方式通过遥控器发出对应的操作指令,以针对遥控器的对应装置来进行操作外,本发明也可直接应用于待操作的电子装置本体上,或是通过电子装置本体间接控制与该电子装置连接的周边装置,其实施状态依设计需求不同而改变,达到任意装置均可通过语音操控的效果O [0067] By design of the present invention, in addition to the remote control or the like is applied, so that a user may issue a corresponding operation instruction through the voice input by way of the remote controller to operate for a corresponding external device to the remote control, the present invention It may also be applied directly to the body of the electronic device to be operated, or indirectly controls the peripheral device connected to the electronic device by the electronic device body, which is designed according to different embodiments state requirements vary, and can be made arbitrarily by means of voice control effect O

[0068] 综上所陈,本发明无论就目的、手段及功效,均显示其迥异于现有技术。 [0068] In summary the Chen, the present invention both in terms of objects, means, and effectiveness, which are quite different from the prior art display. 但是须注意,上述实施例仅为例示性说明本发明的原理及其功效,而非用于限制本发明的范围。 It should be noted that the above embodiments are merely illustrative of the principles and effect of the present invention and is not intended to limit the scope of the invention. 任何熟于此项技艺的人士均可在不违背本发明的技术原理及精神下,对实施例作修改与变化。 Any person can be cooked in this skill in the art without departing the principles and spirit of the present invention, examples of modification and change.

Claims (16)

  1. 1.一种语音辨识方法,应用于一电子装置,其特征在于,该方法包括以下步骤: 取得一声音信号; 转换该声音信号为一数字信号; 自该数字信号中取得多个数值,其中各该数值代表对应的一声波振幅值; 依序比对该多个数值中各该数值相对于至少一相邻数值的大小,以取得一组比对结果值; 判断该组比对结果值是否对应至少一操作指令的其中之一;以及当判断该组比对结果值对应该至少一操作指令的其中之一时,执行对应的该操作指令。 1. A speech recognition method, applied to an electronic device, characterized in that the method comprises the steps of: obtaining an audio signal; converting the sound signal into a digital signal; obtaining a plurality of values ​​from the digital signal, wherein each the sound wave amplitude value corresponding to the representative values; sequentially over each of the plurality of numerical values ​​in the relative size of the at least one value for the adjacent, to obtain a set of comparison result values; determines the set value comparison result corresponds one wherein at least one operating instruction; and when determining the operation command when one of the group should be at least an operation instruction, execute a corresponding comparison result values.
  2. 2.如权利要求1所述的方法,其特征在于,该声音信号通过傅立叶转换而形成该数字信号。 2. The method according to claim 1, wherein the audio signal is formed by Fourier transformation of the digital signal.
  3. 3.如权利要求1所述的方法,其特征在于,通过过滤掉该数字信号中低于一标准值的信号值,以取得该多个数值。 The method according to claim 1, wherein, by filtering out the signal value is lower than a standard value of the digital signal, to obtain the plurality of values.
  4. 4.如权利要求1所述的方法,其特征在于,还包括以下步骤: 检测是否接收一启动信号,以判断是否开始取得该声音信号。 4. The method according to claim 1, characterized by further comprising the steps of: detecting whether a start signal is received, to determine whether to start acquisition of the audio signal.
  5. 5.如权利要求1所述的方法,其特征在于,于判断该组比对结果值是否对应至少一操作指令的其中之一的步骤中,通过比对该至少一操作指令所对应的至少一组参考值与该组比对结果值是否相符或数值差异量在一定范围内,以判断该组比对结果值是否对应该至少一操作指令的其中之一。 5. The method according to claim 1, wherein, in the step of determining which one of the set value comparison result whether the operation instruction corresponding to at least one, at least by a ratio of the at least one operating instruction corresponding It is consistent with the set of reference values ​​of the set value or the comparison result of the amount of the difference value within a certain range, to determine whether the set ratio should be at least one instruction of an operation in which the result values.
  6. 6.如权利要求1所述的方法,其特征在于,将各该数值均与同一侧的一单一相邻数值加以比对,依据各该数值相对于该单一相邻数值的大小状态,以取得对应的该组比对结果值。 6. The method according to claim 1, characterized in that each of the average value to be compared with a single value of an adjacent side of the same, according to each of the state value relative to the size of the adjacent single value, to obtain corresponding to the set value comparison result.
  7. 7.如权利要求1所述的方法,其特征在于,将各该数值同时与位于该数值两侧的该两相邻数值加以比对,依据各该数值相对于该二相邻数值的大小状态,以取得对应的该组比对结果值。 7. The method according to claim 1, wherein each of the values ​​will be simultaneously aligned with the two adjacent sides of the value of the value is in, based on the size of each of the state value relative to the value of the two adjacent , to obtain the corresponding comparison result the set value.
  8. 8.如权利要求1所述的方法,其特征在于,还包括以下步骤: 当判断该组比对结果值未对应该至少一操作指令的其中之一时,检测是否接收一设定信号; 若是,设定该声音信号以对应一新操作指令。 8. The method according to claim 1, characterized by further comprising the steps of: determining when one of the comparison result value should be at least not for an operation wherein the set of instructions, detecting whether a setting signal; if, the audio signal is set to correspond to a new operation command.
  9. 9.一种电子装置,其特征在于,包括: 一声音接收模块,用以接收一声音信号;以及一处理模块,与该声音接收模块电性连接,该处理模块包括: 一转换单元,用以转换该声音信号为一数字信号,并自该数字信号中取得多个数值,其中各该数值代表对应的一声波振幅值;以及一比对编码单元,用以依序比对该多个数值中各该数值相对于至少一相邻数值的大小,以取得一组比对结果值; 通过该处理模块判断该组比对结果值是否对应至少一操作指令的其中之一,以决定是否执行对应的该操作指令。 An electronic device, comprising: a sound receiving module for receiving an audio signal; and a processing module, the receiving module is electrically connected to the sound, the processing module comprises: a converting unit for converting the sound signal into a digital signal, and obtaining a plurality of values ​​from the digital signal, wherein each of the sound wave amplitude value corresponding to the representative value; and a ratio of a coding unit for sequentially comparing the plurality of numerical values the magnitude of each value with respect to at least one adjacent values ​​to obtain a set of comparison result values; determines the set of one of the comparison result whether the value corresponding to at least an operation instruction by the processing module, wherein, to determine whether to perform the corresponding the operation instruction.
  10. 10.如权利要求9所述的电子装置,其特征在于,该处理模块通过傅立叶转换将该声音信号转换成该数字信号。 The electronic device according to claim 9, wherein the processing module converts the digital signal into a sound signal into the Fourier.
  11. 11.如权利要求9所述的电子装置,其特征在于,该转换单元过滤掉该数字信号中低于一标准值的信号值,以取得该多个数值。 11. The electronic device according to claim 9, characterized in that the filtered signal conversion unit is lower than a standard value of the digital signal values ​​to obtain the plurality of values.
  12. 12.如权利要求9所述的电子装置,其特征在于,还包括一启动模块,用以产生一启动信号至该处理模块,以通知该声音接收模块开始取得该声音信号。 12. The electronic device according to claim 9, characterized in that, further comprising a starting module for generating a start signal to the processing module, to inform the sound receiving module to obtain the audio signal starts.
  13. 13.如权利要求9所述的电子装置,其特征在于,还包括一存储模块,用以存储该至少一操作指令所对应的至少一组参考值,其中当该处理模块比对该至少一组参考值的其中之一与该组比对结果值相符或数值差异量在一定范围内时,则判断该组比对结果值对应该至少一操作指令的其中之一。 13. The electronic device according to claim 9, characterized in that, further comprising a storage module for storing at least one set of at least one reference value corresponding to the operation instruction, wherein when the processing module comparing the at least one set one wherein the reference value of the comparison result the set value matches the value of the difference or the amount is within a certain range, it is determined that the set value of the comparison result should be one wherein the at least one operation instruction.
  14. 14.如权利要求9所述的电子装置,其特征在于,将各该数值均与同一侧的一单一该相邻数值加以比对,依据各该数值相对于该单一相邻数值的大小状态,以取得对应的该组比对结果值。 14. The electronic device according to claim 9, characterized in that each of the values ​​are the same with a single value to be adjacent to the side than the right, according to each of the state value relative to the size of the adjacent single value, to obtain the set of values ​​corresponding to the comparison result.
  15. 15.如权利要求9所述的电子装置,其特征在于,将各该数值同时与位于该数值两侧的该二相邻数值加以比对,依据各该数值相对于该二相邻数值的大小状态,以取得对应的该组比对结果值。 15. The electronic device according to claim 9, wherein each of the values ​​will be simultaneously aligned with the two adjacent sides of the value of the value is in, based on each of the relative size of the numerical value of the two adjacent state, to obtain a corresponding comparison result the set value.
  16. 16.如权利要求9所述的电子装置,其特征在于,当该处理模块判断该组比对结果值未对应该至少一操作指令的其中之一时,该处理模块检测是否接收一设定信号,以设定该声音信号以对应一新操作指令。 16. The electronic device according to claim 9, wherein, when the processing module determines the set to be not more than one of the at least one operational instructions wherein the result value, the processing module detecting whether a setting signal, to set the sound signal to correspond to a new operation command.
CN 201110343163 2011-10-27 2011-10-27 Electronic device and method for speech recognition CN103092330B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201110343163 CN103092330B (en) 2011-10-27 2011-10-27 Electronic device and method for speech recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201110343163 CN103092330B (en) 2011-10-27 2011-10-27 Electronic device and method for speech recognition

Publications (2)

Publication Number Publication Date
CN103092330A true CN103092330A (en) 2013-05-08
CN103092330B true CN103092330B (en) 2015-11-25

Family

ID=48205006

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201110343163 CN103092330B (en) 2011-10-27 2011-10-27 Electronic device and method for speech recognition

Country Status (1)

Country Link
CN (1) CN103092330B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104113144B (en) * 2014-06-27 2017-12-05 青岛众海汇智能源科技有限责任公司 Wireless power feeder and a power supply method of multiple loads
CN106296890A (en) * 2016-07-22 2017-01-04 北京小米移动软件有限公司 Mobile terminal unlocking method and device and mobile terminal

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101114448A (en) * 2006-07-28 2008-01-30 佛山市顺德区顺达电脑厂有限公司;神达电脑股份有限公司 Voice identification system
CN100514446C (en) * 2004-09-16 2009-07-15 北京中科信利技术有限公司;中国科学院声学研究所 Pronunciation evaluating method based on voice identification and voice analysis

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020172376A1 (en) * 1999-11-29 2002-11-21 Bizjak Karl M. Output processing system and method
US8155965B2 (en) * 2005-03-11 2012-04-10 Qualcomm Incorporated Time warping frames inside the vocoder by modifying the residual

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100514446C (en) * 2004-09-16 2009-07-15 北京中科信利技术有限公司;中国科学院声学研究所 Pronunciation evaluating method based on voice identification and voice analysis
CN101114448A (en) * 2006-07-28 2008-01-30 佛山市顺德区顺达电脑厂有限公司;神达电脑股份有限公司 Voice identification system

Also Published As

Publication number Publication date Type
CN103092330A (en) 2013-05-08 application

Similar Documents

Publication Publication Date Title
CN101162398A (en) Arbitrarily signal generating device
CN102005070A (en) Voice identification gate control system
CN104714741A (en) Method and device for touch operation
CN101453598A (en) Electronic device and method capable of sound effect regulation according to customer location
CN103024630A (en) Volume regulating method of first electronic equipment and first electronic equipment
US20150310878A1 (en) Method and apparatus for determining emotion information from user voice
CN102637353A (en) System and method for home appliance wireless remote control based on mobile terminal
CN101345819A (en) Speech control system used for set-top box
US20140129235A1 (en) Audio tracker apparatus
US20140028826A1 (en) Voice recognition method and apparatus using video recognition
US20150040740A1 (en) Sampling device and sampling method
CN101447119A (en) Household appliance and control method thereof
CN203232565U (en) Multifunctional remote network remote controller
CN102722369A (en) Information processing terminal, information processing method and application of method in intelligent television or set-top box of intelligent television
CN202736280U (en) Remote control fast matching device
US20140348336A1 (en) Electronic apparatus and method for activating specified function thereof
CN102722242A (en) Method and device for operating touch screen of electronic equipment
CN103049186A (en) Electronic device with sound control button function and sound control button method
Haleem Voice controlled automation system
CN2681491Y (en) Voice demander for television
CN103886236A (en) Acoustic control screen unlocking method and mobile terminal
CN102025834A (en) Mobile terminal voice operation method and device
CN104850222A (en) Instruction recognition method and electronic terminal
US20080100747A1 (en) Voice-controlled TV set
CN202160214U (en) Device for converting mobile phone application dynamic display effects by voice control

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model