JP2006165846A - Portable telephone device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a portable telephone device for easily retrieving desired recording data. <P>SOLUTION: At the time of recording speech, a record data generating part 107 separately records an incoming speech voice at a user side and an outgoing speech voice at the other side, and generates the record data at a set bit rate. At the time of retrieving speech recording, words or phrases which seem to be present during a speech with the voice of a user are inputted from a microphone 109 as a voice, and the voice is converted into waveform, and a voice waveform for retrieval is generated by a voice waveform generating part 102. A waveform retrieving part 112 retrieves recording data having waveform almost matched with voice waveform from the incoming speech voice of the narrowed-down recording data. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a mobile phone device.

  2. Description of the Related Art Conventionally, some communication devices such as mobile phone devices have a call recording search function for recording a call content and easily searching for and reproducing a call content with a necessary communication partner.

  Examples of this type of communication device include a mobile phone and computer software disclosed in Patent Document 1. This mobile phone includes a call recording unit that records call contents, a database that stores call contents for each communication partner, a database control unit, and the like.

  The call recording unit records the call with the communication partner set in advance at the time of the call for each communication partner, and the database control unit registers by the communication partner by specifying the communication partner by specifying the telephone number in the registered phone book. To control. Therefore, when searching for and playing back the content of a call with a necessary communication partner, the user designates the telephone number of the necessary communication partner, reads the telephone number registered in the database, and performs communication corresponding to the telephone number. The registered contents can be played from the other party's file.

  Moreover, there is a call recording device disclosed in Patent Document 2 as a call recording device that can easily search for desired call contents. This call recording device automatically records the call contents of the telephone device, and allows the necessity of storing the call contents to be selected after the call. If it is not necessary to save the call content, the call content is deleted. If it is necessary to save the call content, the call content is saved and a title explaining the call content is recorded. As a result, the recorded call contents can be searched by listening to the contents of the title at the time of reproduction.

However, the conventional cellular phone device has a problem that it becomes difficult to search for a call content to be obtained from a call partner or a call date and time when the recording data increases. In addition, it is difficult to perform a search based on the content of a call such as “patent”.
Japanese Patent Laid-Open No. 2004-187072 JP-A-8-149226

  The present invention has been made in view of the above-described conventional circumstances, and an object of the present invention is to provide a mobile phone device that can easily search for desired recording data by calling voice.

  The cellular phone device according to the present invention includes a recording data generation means for recording a voice during a call to generate recording data, a recording data storage means for storing the recording data, and a search for inputting a voice of a user for search. Audio search means and search means for searching for recorded data that substantially matches the voice waveform of the input search voice with reference to the recording data storage means.

  With this configuration, it is possible to easily search for desired recording data by searching for recording data with one's own voice.

  In the mobile phone device of the present invention, the recording data generating means generates recording data by dividing an uplink call that is a user's voice and a downlink call that is a voice of a call partner.

  With this configuration, the uplink call and the downlink call are not mixed, and the voice waveform of the uplink call is not affected even when the noise on the downlink call side is large. Furthermore, since the user's voice is used for the search, the recognition rate at the time of the voice search can be increased.

  In the mobile phone device of the present invention, the recording data generation means generates the recording data for the uplink call and the downlink call at different bit rates.

  With this configuration, it is possible to set the uplink call voice necessary for the search to a high bit rate and suppress the downlink call voice to a bit rate that can be sufficiently heard. As a result, the capacity of the recording data storage means required for recording can be reduced as compared with the case where recording data is generated at the same bit rate.

  Further, in the cellular phone device of the present invention, the recording data generation means adds silence identification information indicating a silent part during a call to the recording data, and the search means includes the silence part indicated by the silence identification information. Search the recorded data except the part.

  With this configuration, the search time can be reduced by skipping the silent portion, and the silence identification information can be used as a search start point, thereby improving the search efficiency.

  According to the present invention, it is possible to provide a mobile phone device that can easily search for desired recording data from a large amount of data that is noisy like a phone call.

  FIG. 1 is a block diagram showing a configuration of a mobile phone device according to an embodiment of the present invention. As shown in FIG. 1, the mobile phone device 100 includes a transmission / reception unit 101, a voice waveform generation unit 102, an operation unit 103, a display unit 104, an upstream voice detection unit 105, a downstream voice detection unit 106, and a recording. A data generation unit 107, a codec 108, a microphone 109, a speaker 110, a partner information generation unit 111, a waveform search unit 112, a database 113, a central processing unit 114, and a silent part detection unit 115 are provided.

  The transmission / reception unit 101 performs transmission / reception via an antenna (not shown). The voice waveform generation unit 102 converts the input voice into a voice waveform. The operation unit 103 performs various setting operations of the mobile phone device 100. The display unit 104 displays characters and various operating states of the mobile phone device 100.

  The uplink voice detection unit 105 detects the uplink voice, that is, the user's voice input from the microphone 109. The downlink voice detecting unit 106 detects the downlink voice, that is, the voice on the other party side received from the transmission / reception unit 101. The recorded data generation unit 107 combines the voice data detected by the upstream voice detection unit 105 and the downstream voice detection unit 106 with the call partner information. The microphone 109 is used to input an instruction voice or a call voice from the user, and the speaker 110 outputs a call voice from the call partner side. The codec 108 performs A / D conversion, D / A conversion, and encoding / decoding processing on audio data input from the microphone 109 and audio output to the speaker 110.

  The other party information generation unit 111 generates the other party information using the call start time and the other party telephone number. The waveform search unit 112 searches the recorded data for a voice waveform that matches the voice waveform input for the search. The database 113 stores recorded data including a call start time, a call end time, a call partner telephone number, and the like. The central processing unit 114 performs control and signal processing of each unit of the mobile phone device 100. The silent part detector 115 detects a silent part during a call.

  The cellular phone device 100 according to the present embodiment can set (1) whether or not to record a call voice, and has a function of automatically recording a call when it is done. In addition, (2) a function for separately recording the voice of the user's incoming call and the other party's outgoing call, and (3) a silent flag is set as a mark in the silent part. (4) Function for setting / changing the bit rate during recording of uplink and downlink calls, (5) Function for converting input voice into voice waveform, (6) Input voice from microphone, A function for searching for a call voice having an uplink voice waveform that substantially matches the voice waveform, (7) a function for playing back the searched call together with the uplink voice and the downlink voice, and (8) a call start time and a call end time for the recorded data. In addition, it has a function of providing other party information.

  The above functions are realized according to the control of the central processing unit 114. For example, through the operation unit 103, the user sets whether to record a call and sets / changes the bit rate when recording an uplink call and a downlink call.

  In addition, when the call processing is set to be performed, the central processing unit 114 has an incoming or outgoing call. When the call starts, the central processing unit 114 extracts the call start time and the other party's telephone number from the other party information generation unit 111, and records data. The generation unit 107 generates recording data having both call voice and call partner information, and controls each unit to automatically record.

  Further, when recording the call voice, the central processing unit 114 separately records the voices of the uplink call and the downlink call, and sets a flag serving as a mark as silence identification information in each silence part. Also, recording ends at the same time as the call ends. At the end of the call, call attribute data such as importance, erasure prohibition setting, secret setting, etc. is automatically assigned to the call voice by the input of the operation unit 103 from the user, and stored in the database 113 as recording data. FIG. 2 is an example showing the structure of the recording data of the mobile phone device according to the embodiment of the present invention.

When searching for the desired recording data from the stored recording data, the user operates the operation unit 103 of the mobile phone device 100 to place the voice input in a standby state, and a word or phrase that seems to have been in a call. Is input to the microphone 109 as voice, and the voice waveform generation unit 102 converts the voice into a waveform to generate a voice waveform. The waveform search unit 112 searches the recorded data recorded in the database 113 for recorded data that is substantially the same as the input voice waveform, generates a substantially matched call list, and displays the list on the display unit 104. The user selects and reproduces the recorded data that seems to be the target call from the substantially matched call list.
FIG. 3 is a diagram showing a method for searching recorded data of the mobile phone device according to the embodiment of the present invention. As a search method, a recorded call is searched from the beginning of the call, and a portion where the waveforms are substantially matched ((1) in the figure) is searched. When the call is long or the user's voice used for the search is If it is known that the call is in the second half of the call, there is a method in which the user sets the search location and shortens the search time ((2) in the figure). In addition, since the call voice is processed as data, it is possible to search backwards from the voice, that is, the time series when the voice is recorded ((3) in the figure). When the voice to be searched is at the end of the call, the search time can be shortened by searching from the reverse. Therefore, when the call is long or when the voice to be searched is known in advance in the call, the search time can be shortened.

  In addition, the mobile phone device 100 according to the present embodiment uses the user's call voice input from the microphone 109 and the other party's call voice from the speaker 110 to facilitate searching after recording. It is possible to record a certain downstream call separately. This makes it possible to find a target call with high accuracy from the recorded data. For example, if you record upstream and downstream calls as one recording data and use that voice for search, when the upstream and downstream voices are emitted at the same time, the waveform of the voice you uttered will change, and that part will be searched When used for, the voice recognition rate decreases. By recording the uplink and downlink voices separately, it is possible to prevent noise from the downlink call voice and perform an efficient search. In addition, since it has a function that can change the bit rate at the time of recording of the uplink and downlink calls, the capacity of the database 113 is taken into consideration, and the uplink call voice necessary for the search is set to a high bit rate. The bit rate can be suppressed to a level that can be heard sufficiently. As a result, the capacity of the database 113 required for recording can be reduced as compared with the case where all are recorded at the same bit rate. The database here refers to a memory card, a magnetic disk, an optical disk, etc., and may be fixed to the mobile phone device or removable. In the case of a removable database, the memory capacity can be easily increased by replacing the memory card, and the database can be removed so that it is not accidentally deleted when important calls are saved. Increased safety and safety.

  With respect to the mobile phone device 100 configured as described above, the operation during recording will be described using the flowchart shown in FIG.

  First, in the standby state (step S201), the presence / absence of on-hook at the time of outgoing or incoming is detected (step S202).

  If there is no on-hook in step S202, the process returns to step S202, and the presence / absence of the on-hook at the time of outgoing or incoming is detected again.

  On the other hand, if on-hook is detected in step S202, it is determined whether or not to start recording by setting whether or not to automatically record a call set in advance by the user. Here, when the user sets to automatically record a call, the other party information generation unit 111 reads the call start time and the other party telephone number, and generates them as the other party information (step S203). Then, the recording data generating unit 107 creates recording data having the other party information and starts recording (step S204).

  In order to facilitate the search, the call voice is recorded separately for the uplink call voice that is the user's call input from the microphone 109 and the downlink call voice that is the call of the other party received from the transmission / reception unit 101. Therefore, when the upstream speech detection unit 105 detects upstream speech input and detects upstream speech input, the upstream processing unit 114 notifies the central processing unit 114 that upstream speech input has been detected. It is determined whether there is an uplink voice input based on the presence or absence of the uplink voice input detection notification from the detection unit 105 (step S205).

  If it is determined in step S205 that an uplink voice has been input, a flag that is silence identification information is inserted (step S206), and the uplink voice is recorded (step S207). Here, when recording the uplink call voice, the call voice input from the microphone 109 is recorded by the codec 108 at the set bit rate at the time of the uplink call recording, and the voice waveform generation unit 102 converts the call voice into a voice waveform. After being converted, the database 113 records the recording data.

  The silent part detection unit 115 detects the silent part. When the silent part is detected, the central processing part 114 notifies the central processing unit 114 that the silent part has been detected, and the central processing part 114 detects the silent part from the silent part detection unit 115. It is determined whether or not a silent part has been detected based on the presence or absence of notification (step S208).

  If it is determined in step S208 that no silence has been detected, the process returns to step S207 to record the uplink voice again.

  On the other hand, if it is determined in step S208 that a silent part has been detected, silence identification information is inserted (step S209), and it is determined whether the call has ended (step S210).

  If it is determined in step S210 that the call has not ended, the process returns to step S205 to determine again whether there is an uplink voice input.

  On the other hand, if it is determined in step S210 that the call has ended, the other party information generation unit 111 adds the call end time to the other party information, and the recorded data generated by the database 113 and the user operates the operation unit 103. The input importance, erasure prohibition, secret, etc. are recorded as attribute information, and the recording is stopped (step S216).

  If the central processing unit 114 determines in step S205 that there is no upstream voice input, it determines whether there is downstream voice (step S211). Here, when the downlink voice detection unit 106 detects the downlink voice input and detects the downlink voice input, it notifies the central processing unit 114 that the downlink voice input has been detected, and the central processing unit 114 detects the downlink voice input. Based on the presence or absence of a downstream voice input detection notification from 105, it is determined whether there is a downstream voice input.

  If it is determined in step S211 that there is no downlink voice, the process proceeds to step S210, and it is determined whether the call is finished.

  On the other hand, if it is determined in step S211 that there is a downlink voice, silence identification information is inserted (step S212), and the downlink voice is recorded (step S213). Here, when recording the downlink call voice, the voice waveform generation unit 102 converts the call voice into a voice waveform and records at the bit rate at the time of the downlink call recording in which the codec 108 is set, as in the case of recording the uplink call voice. The recorded data generation unit 107 records the call voice as recorded data in the database 113.

  The silent part detection unit 115 detects the silent part. When the silent part is detected, the central processing part 114 notifies the central processing unit 114 that the silent part has been detected, and the central processing part 114 detects the silent part from the silent part detection unit 115. It is determined whether or not a silent part has been detected based on the presence or absence of notification (step S214).

  If it is determined in step S214 that no silence has been detected, the process returns to step S213 to record the downlink voice again.

  On the other hand, if it is determined in step S214 that a silent part has been detected, silence identification information is inserted (step S215), and the process proceeds to step S210 to determine whether the call has ended.

  When there is no room in the capacity of the database 113, the priority order may be determined based on the attribute information of the recording data, and the newest call may be recorded by automatically deleting the priority order from the lowest order.

  As described above, the mobile phone device of the present embodiment separately records the voice of the uplink call that is the user's call and the downlink call that is the call of the other party during recording, and the uplink voice voice and the downlink voice voice. Each has a function of detecting a silence part, and has a function of installing silence identification information at the beginning and end of the silence part. As a result, it is possible to subdivide the call, and the silent part can be omitted even when searching, and the silence identifying information is used as a search starting point, so that the search time is shortened. In addition, by setting different bit rates when recording an upstream call and a downstream call, the capacity of the database 113 required for recording and storing can be reduced compared to the case where all calls are recorded at the same bit rate.

  Next, the operation when searching for specific call data from the recorded data recorded in the procedure shown in FIG. 4 will be described based on the flowchart shown in FIG.

  When the search is started, the central processing unit 114 determines whether the search is a conditional search (step S301). Here, the search is performed by voice waveform. However, since it is inefficient to search for a huge amount of all calls, it is possible to narrow down the calls in advance according to conditions such as the call partner, call date and time, and rough call time. Is possible. When the user searches the recording data, the user inputs conditions for narrowing down the call recording search, and the mobile phone device 100 reads and searches the call time, call time, and destination information recorded in the call recording, thereby making a call. It is possible to narrow down.

  If it is determined in step S301 that the search is conditional, a search condition is set by the user (step S302). Search conditions that can be set include, for example, a call partner, a call date and time, and a rough call time.

  If it is determined in step S301 that the search is not a conditional search, or after narrowing down according to the search condition set in step S302, a search using a speech waveform is performed. The user puts the voice input into a standby state by operating the operation unit 103 of the mobile phone device 100, and inputs a word or phrase that seems to have been in a call to the microphone 109 as a voice. 102 records the voice as a waveform and generates a voice waveform for search (step S303).

  The waveform search unit 112 searches for the recording data having a waveform that substantially matches the voice waveform generated in step S303 from the uplink voice of the narrowed recording data (step S304). Here, when the waveform search unit 112 finds recording data having a waveform that substantially matches the generated voice waveform, the waveform search unit 112 notifies the central processing unit 114 that a search waveform and a substantially matching waveform have been found.

  The central processing unit 114 determines whether there is a waveform that substantially matches the speech waveform for search based on the presence / absence of a notification that the waveform search unit 112 has found a substantially matched waveform (step S305).

  In step S305, when the central processing unit 114 receives the notification that the substantially matching waveform is found from the waveform searching unit 112 and determines that there is a substantially matching waveform, the central processing unit 114 reads information on the recording data that substantially matches from the database 113. Then, the recorded data is added to the substantially matched call list, an approximate matched call list is created (step S306), and it is determined whether or not all calls have been searched (step S307).

  If it is determined in step S307 that all the recorded data has not been searched, the process returns to step S304 and substantially matches the audio waveform generated in step S303 from the uplink audio of the recorded data that has been narrowed down again. Search for calls with waveforms to play.

  On the other hand, if it is determined in step S307 that all the recorded data have been searched, it is determined whether the number of recorded data substantially matching the voice waveform generated in step S303 is greater than 0 (step S308).

  If it is determined in step S308 that the number of substantially matched calls is not greater than 0, it is determined that a waveform that substantially matches the voice input waveform generated in step S303 is not found, and it is determined whether to search again ( Step S309) If it is determined to search again, the process returns to Step S301 and the search is started again.

  On the other hand, if it is determined in step S309 that the search is not performed again, the process proceeds to step S314, and the process ends.

  If it is determined in step S308 that the number of recorded data that substantially matches the voice waveform generated in step S303 is greater than zero, the display unit 104 displays a substantially matching call list from which the user can The item that seems to be recorded data is selected (step S310). By displaying the substantially matched call list on the display unit 104, the user is notified that the recorded data has been found, and the call time and the other party's telephone number that the recorded data has. In addition, if the phone number recorded in the recorded data found by the search is registered in the phone book of the mobile phone, the name of the other party in the phone book and other information registered in the phone book are displayed. It is displayed on the display unit 104.

  In order to confirm the selected recording data, the central processing unit 114 controls to reproduce the selected recording data (step S311). The call recording / reproduction here reproduces the uplink call voice and the downlink call voice between the n-th silence identification information before and after the voice waveform for the search and the portion substantially coincident. N at this time is determined by the setting of the user. It is possible to facilitate the confirmation by reproducing only the nth silent identification information before and after, and to shorten the search time as a result.

  The user confirms whether or not the reproduced recording data is the intended one (step S312). The user determines whether the recorded data is the intended one by listening to the reproduced recording data. If the recorded data is the intended one, the user operates the operation unit 103 to reproduce the recorded call recording. Is sent to the central processing unit 114, and the central processing unit 114 confirms whether the recorded data reproduced by the notification from the operation unit 103 is the intended one.

  If it is determined in step S312 that the selected call is not the intended one, the process returns to step S310 to select recording data again.

  If the selected recording data is the target in step S312, the recording data is reproduced (step S313). Even during playback, it is possible to play back the entire beginning of the corresponding call, but by using silence identification information, by specifying the range before and after the match, the target call can be easily set even for long calls. Can be searched and played.

  If the target call is not found even if all the call recordings in the substantially matched call list are played back in steps S310 to 312, the process returns to step S303, and the search is performed again by inputting voice. The user can decide whether to end the search.

  As described above, after narrowing down according to the search condition, an uplink call recording that substantially matches the voice waveform for user search is searched. As a result, it is possible to easily search the recorded data of the desired call content with his / her voice.

  According to such a mobile phone device of the embodiment of the present invention, during the recording, the uplink voice as the user's call voice and the downlink voice as the other party's call are separately recorded, and the uplink voice and Each of the down-call voices has a function of detecting a silence part, and silence identification information is set at the beginning and end of the silence part. After narrowing down according to the search condition at the time of the search, the upstream recording voice that substantially matches the user's voice waveform is searched, and when the target call voice is found, the call voice is reproduced. As a result, it is possible to easily search the recorded data of the desired call content with his / her voice.

  In the above description, the example in which the silence identification information is installed in the voice data has been described. However, in some cases, the silence identification information installation time of each of the uplink call voice and the downlink call voice and the time from the start of the call are listed. It is also possible to install by recording.

  INDUSTRIAL APPLICABILITY The present invention has an effect that it is possible to easily search for desired recording data from contents by voice, and is useful for a mobile phone device or the like.

The block diagram which shows the structure of the mobile telephone apparatus which concerns on embodiment of this invention. The figure which shows the structure of the recording data of the mobile telephone apparatus which concerns on embodiment of this invention. The figure which shows the search method of the recording data of the mobile telephone apparatus which concerns on embodiment of this invention. The flowchart which shows the operation | movement at the time of recording of the mobile telephone apparatus which concerns on embodiment of this invention. The flowchart which shows the operation | movement at the time of the search of the mobile telephone apparatus which concerns on embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Cellular phone apparatus 101 Transmission / reception part 102 Voice waveform generation part 103 Operation part 104 Display part 105 Upstream voice detection part 106 Downstream voice detection part 107 Recording data generation part 108 Codec 109 Microphone 110 Speaker 111 Partner information generation part 112 Waveform search part 113 Database 114 Central processor 115 Silent part detector

Claims (4)

Recording data generation means for recording voice during a call and generating recording data;
Recording data storage means for storing the recording data;
A search voice detection means for inputting a search user's voice;
Search means for searching for recorded data substantially matching the voice waveform of the input search voice with reference to the recording data storage means;
A mobile phone device comprising: The mobile phone device according to claim 1,
The recorded data generation means is a mobile phone device that generates recorded data by dividing an uplink call that is a user's voice and a downlink call that is a call partner's voice. The mobile phone device according to claim 2, wherein
The recording data generation means is a mobile phone device that generates recording data of the uplink call and the downlink call at different bit rates. The mobile phone device according to any one of claims 1 to 3,
The recorded data generation means adds silence identification information indicating a silent part during a call to the recorded data, and the search means searches for recorded data in a portion excluding the silent part indicated by the silence identification information. Mobile phone device.

Images