JPH11184887A - Device for storing and retrieving digital information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate the necessity to provide any exclusive operation button by easily detecting a desired data position even under an environment in which voice recognizing precision is deteriorated, or even when the same voice index is registered at plural positions. SOLUTION: At the time of retrieving a voice index which is made coincident with a voice from a microphone 21 in a voice index storage memory 5, when the voice index with prescribed similarity or more is extracted, they are sequenced in the order of similarity, and a desired identification code is selected according to the above, and digital information is retrieved based on the identification code. Therefore, a desired data position can be easily detected by preventing any repeated processings due to a recognition error even under an environment in which voice recognizing precision is deteriorated die to a background noise, or even when the same voice index is registered at plural positions. The instruction of an operation related with a retrieval processing by noise input is operated by continuously pressing a a basic storage and reproduction operation button for a prescribed time of longer.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for storing and reproducing digital information such as music and images, and adding a voice index (identification code) to the digital information so as to generate a digital signal based on a voice instruction. The present invention relates to a digital information storage and retrieval device having a function of retrieving information.

[0002]

2. Description of the Related Art In recent years, rapid development of technology in both hardware and software, such as downsizing and increasing the capacity of memory and increasing the efficiency of data compression technology, has enabled information such as voice, audio, still images, and moving images. It is stored in large quantities as digital data,
It is possible to make the storage / playback device capable of playback extremely small. For example, if the voice quality is equivalent to that of the telephone band, it can be efficiently compressed by the CELP method.
CS-A standardized by "T Recommendation G.729"
With CELP, the bit rate can be as low as 8 Kbps. Also, "Japanese Electronics April 2, 1997
The 1st issue introduces the TwinVQ standard which can compress even CD-quality audio data to about 80 Kbps. For example, if a 64 Mbit memory is used, several pieces of music data for one CD can be stored.
In this way, it has become possible to configure a storage / playback apparatus in a small size such as a PCMCIA standard card. However, while the portability of the device is improved due to the miniaturization of the device main body, the operation buttons and the display unit are also reduced in size due to the space, and there is a problem that the operability is deteriorated. This problem is particularly serious when complicated operations such as searching for a desired data position and cueing are required. As one means for solving such a problem, it is effective to use a voice recognition technique that does not require complicated button operations. Therefore, a storage / reproducing apparatus using the above-described voice recognition technology is disclosed in, for example, Japanese Patent No. 2639387,
No. 3,389,872. These storage / playback apparatuses add a voice index to the beginning of music, voice data, etc., and input a voice corresponding to the desired index during playback, thereby searching for an index position that matches the voice. Then, playback from that position is performed automatically.

[0003]

However, with the current speech recognition technology, it is impossible to perform 100% accurate recognition under the normal use environment without being affected by background noise or the like. Therefore, the recognition accuracy of the voice is reduced depending on the surrounding environment. Even in the cueing operation as described above, there are many malfunctions in actual use, and the user has to input the voice each time and repeat the operation. Also, there is no guarantee that recognition will be accurate even if the process is repeated. In addition, the above-mentioned patent No. 26
In the invention according to 39387, the recognition rate is increased by, for example, performing registration five times with the same voice. However, the registration operation is troublesome for the user, and erroneous recognition due to a change in the surrounding environment is required. Can not. In addition, when exactly the same audio index is registered for different music or audio data, they cannot be distinguished, so that they cannot be operated as desired. Therefore, for example, the above-mentioned Patent No. 263
In the invention according to No. 9387, each voice index is always differentiated by, for example, pronouncing a number at the end of the voice index, but this is very troublesome for the user. Further, in the above-mentioned conventional storage / reproduction device, operation buttons used for registering, inputting, and retrieving the audio index are provided exclusively. However, in addition to the basic reproduction and recording buttons, these operation buttons are further provided. It is very difficult to provide such a small device such as a card type as described above due to space restrictions. SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to perform a data search or the like by voice, and to be performed in an environment where voice recognition accuracy is reduced such as when there is background noise. However, even when the same voice index is registered at a plurality of positions, it is possible to easily detect a desired data position, and it is necessary to provide a dedicated operation button related to the operation by the voice. The object of the present invention is to provide a digital information storage / retrieval apparatus without any problem.

[0004]

To achieve the above object, the present invention provides a storage means for storing digital information, a voice input means for inputting voice, and a voice input by the voice input means. Identification code storage means for storing the digital information stored in the storage means as an identification code indicating the position on the storage means; voice input by the voice input means and identification code stored in the identification code storage means; And a search means for searching for the position of the digital information on the storage means on the basis of the above, wherein the voice input is selected from the identification codes stored in the identification code storage means. Extracting one or a plurality of sounds similar to the voice input by the means,
A digital information storage and retrieval device comprising: similar identification code ordering means for performing a predetermined ordering; and identification code selection means for selecting a desired identification code in accordance with the ordering performed by the similarity identification code ordering means. Have been.

[0005] Further, it is possible to comprise a reproducing means for reproducing the digital information searched by the searching means. Examples of the predetermined ordering method include, for example, a method performed based on the degree of similarity between each identification code and the voice input from the voice input means, and a method based on the storage position indicated by each identification code on the storage means. There is a method to do so. Further, the operation of storing the identification code of the identification code storage means based on the voice input means, the instruction of the ordering operation by the similar identification code ordering means based on the voice input means, and the operation of the identification code selection means Of the digital information can be realized by a predetermined operation on the storage / reproduction input means for inputting the instruction relating to the storage / reproduction of the digital information, input means such as dedicated buttons for processing by voice input can be provided. Since there is no need to provide them, it is possible to effectively utilize space even in a small device. If the processing by the voice input is performed by, for example, continuously operating the storage / reproduction input means for a certain period of time or more, it is easy to provide the storage / reproduction input means with a plurality of functions.

[0006]

According to the digital information storage and retrieval device of the present invention, the position of digital information on the storage means is determined based on the voice input by the voice input means and the identification code stored in the identification code storage means. At the time of retrieval, one or a plurality of speeches similar to the speech input by the speech input means are extracted from the identification codes stored in the identification code storage means by the similar identification code ordering means. In order. Then, a desired identification code is selected by the identification code selection means in accordance with the ordering by the similar identification code ordering means, and digital information is searched based on the identification code. Therefore,
For example, a desired data position can be easily detected even in an environment where the voice recognition accuracy is reduced due to background noise or the like, or when the same voice index is registered at a plurality of positions. In addition, an instruction of an operation related to a search process by voice input, such as an operation of storing an identification code and an ordering operation by the similar identification code ordering means, is performed by a storage / reproduction input means for inputting an instruction relating to storage / reproduction of the digital information. For example, it is performed by performing continuous operation for a predetermined time or more. Therefore, it is not necessary to separately provide an input means dedicated to processing by voice, so that space can be effectively used even with a small device.

[0007]

Embodiments and examples of the present invention will be described below with reference to the accompanying drawings to provide an understanding of the present invention. The following embodiments and examples are mere examples embodying the present invention, and do not limit the technical scope of the present invention. Here, FIG. 1 is a block diagram showing a schematic configuration of an information storage / reproduction device A1 according to an embodiment of the present invention, FIG. 2 is a schematic diagram showing operation buttons of the information storage / reproduction device A1, and FIG. FIG. 4 is an explanatory diagram showing a configuration example of information stored in an information address management memory, a voice index storage memory, an information body storage memory, and a voice index search history memory in the playback device A1, and FIG. FIG. 5 is a flowchart showing a processing procedure of a voice index registration operation at the time of information reproduction, and FIG. 5 uses a “pause button”.
FIG. 6 is a flowchart showing a procedure for registering an audio index in the middle of information, FIG. 6 is a flowchart showing a procedure for an information search operation using a “forward skip button”, and FIG. It is a flowchart which shows the processing procedure of an information search operation. The information storage / reproduction device A1 according to the present embodiment is a device for storing / reproducing audio information (an example of digital information) such as music, and has a schematic configuration as shown in FIG.

As shown in FIG. 1, the information storage / reproduction device A1 includes a control unit 14 for centrally controlling the operation of each unit described later.
It is composed mainly of. The audio information stored in the information storage / reproduction device A1 is input via the information writing interface 13 and is converted into a data-compressed bit stream as an information body storage memory 1 comprising a semiconductor memory.
It is stored (recorded) in 0 (see the lower left diagram in FIG. 3). FIG.
1 shows an example in which data of one music piece is connected to a continuous area in the memory and stored for simplicity. As the information writing interface 13, for example, a PCMCIA interface or the like is suitable for a card type device, but other than this, serial communication, cable communication such as parallel I / O, infrared communication such as IrDA, wireless communication, etc. ， Various interfaces are available. Management information such as the storage address of each piece of information in the information body storage memory 10 is stored in the information address management memory 9. The information address management memory 9
As shown in the upper left part of FIG. 3, the management information stored in the storage unit includes accompanying information such as a song name and an artist name of each information,
A start address (S1, S2) indicating where each piece of information (song) is stored in the information body storage memory 10.
) And end addresses (E1, E2, etc.).

When reproducing the audio information in the information body storage memory 10, first, the audio information that has been compression-encoded is decoded by the information reproducing unit 11, and is, for example, 16 bits.
It is restored to digital data sampled in stereo. The digital data is converted from an analog signal by the D / A converter 12 and then output from an external output device 22 such as a headphone or a speaker. A voice registered as a voice index (identification code) indicating an arbitrary position of voice information stored in the information main body storage memory 10 and an input voice for searching the voice index are a microphone 21 (voice input means). , And digitized by the A / D converter 1, and the voice detector 2 determines whether or not voice is included in the digitized signal. Various known techniques can be applied, such as a simple determination based on a pattern of volume fluctuation, a method based on parameters extracted from the signal such as zero crossing and periodicity of the signal, and the like. The voice detected by the voice detection unit 2 is converted to a voice parameter conversion unit 3
Some speech parameters P11, P12, P
21, are converted. As the parameters used here, LPC cepstrum, mel cepstrum, and the like generally used in speech recognition are effective. The parameters are usually quantized in order to reduce the amount of memory, and it is already known that modeling with an HMM (Hidden Markov Model) is also effective.

Subsequently, when the input voice is to be registered as a voice index, the number of the voice information being recorded (or being reproduced) at that time, the position address of the recording pointer (or reproduction pointer) at that time, The voice parameters (P11, P12, etc.) are stored as voice indexes in the voice index storage memory 5 (identification code storage means) (see the upper right diagram in FIG. 3). on the other hand,
If the input speech is for searching for a speech index, the speech matching similarity calculation unit 4 (similar identification code ordering means) converts the speech parameters converted by the speech parameter conversion unit 3 (input speech parameters). ) And each voice parameter (referred to as a registered voice parameter) stored in the voice index storage memory 5. As a result, one or more voice index numbers in which registered voice parameters whose similarity with the input voice parameter is higher than a predetermined threshold are registered are extracted. For example, ordering is performed in the order of the similarity, and voice index is performed. Search history memory 8
(See voice index search history memory in FIG. 3). The recording and reproduction of the audio information and the registration and search of the audio index described above are performed by the information storage / reproduction device A.
The operation is started by pressing an operation button 15 provided on one main body. The operation button 15 is, as shown in FIG.
It is composed of only basic buttons (storage / playback input means) necessary for recording / playback of audio information, such as recording, playback, pause, and skip. Operations such as registration and search of a voice index are realized by continuously pressing the basic operation button for a predetermined time or more. Button operation detector 6
Detects the state of the operation button 15 and notifies the control unit 14 by interrupt or I / O, for example. Timer 7
Measures the time that the operation button 15 is kept pressed. The control unit 14 determines which operation (the basic operation or the operation related to the voice index) the button operation is defined by the operation button based on the time measured by the timer 7.

Next, the operation of the information storage / reproducing apparatus A1 will be described in detail with reference to the flowcharts shown in FIGS. [Registration operation of audio index at the time of information reproduction] The "reproduction start button" shown in FIG. 2 is also used as the start button of the audio index registration operation at the time of information reproduction. As shown in FIG. 4, when the "reproduction start button" is pressed (step S1), the operation is detected by the button operation detection unit 6, and the control unit 14 instructs the information reproduction unit 11 to store the information in the information body storage memory 10. The reproduction of the audio information (music) is instructed, and the reproduction of the music starts (step S2). Up to here, the normal reproduction operation when the “reproduction start button” is pressed. The control unit 14 simultaneously activates the audio detection unit 2 to start detection of audio in the input signal from the microphone 21 (step S3), and activates the timer 7 so that the "playback start button" is activated. The measurement of the pressed time is started (step S4). When the "playback start button" which has been pressed is released (step S5), it is determined whether or not the time measured by the timer 7 at that time is equal to or longer than a predetermined threshold (for example, 0.5 seconds). (Step S6). Here, if the measurement time is less than the threshold, no special operation is performed, and the reproduction operation is continued. That is, when the "reproduction start button" is pressed for a short time, only a normal reproduction operation is performed.

On the other hand, if the measured time is equal to or longer than the threshold, that is, if the "playback start button" is continuously pressed for a predetermined time or longer, the control unit 14 determines that the requested operation is performed in accordance with the audio index. It is determined that the operation is a registration operation, and the following processing is performed. That is, it is determined whether or not the voice is detected by the voice detection unit 2 (step S7). If the voice is detected, the voice is converted into some parameters by the voice parameter conversion unit 3 and the current playback is performed. It is registered in the voice index storage memory 5 as a voice index indicating the beginning of the middle tune (step S8). The audio index registered in this case is registered so as to indicate the beginning of the music in order to find the beginning of the music. As shown in 2,
The song number and voice parameters P21, P22, P2
.. Are registered, and an intermediate address indicating an intermediate address of the music is not set. On the other hand, if no sound is detected in step S7, that is, if the "playback start button" is continuously pressed for a predetermined time without inputting a sound, the beginning of the currently played back music is displayed. The registered voice index shown is deleted (step S
9). For example, the voice index No. in the upper right diagram of FIG.
2 is deleted. As described above, the "playback start button" allows not only normal playback operation but also instructions for voice index registration / deletion operation, and there is no need to provide a special dedicated button. Is improved.

[Sound index registration operation at the time of recording information] The "record button" shown in FIG. 2 is also used as a start button of the audio index registration operation at the time of recording information. Note that the operation of registering the voice index using the “record button” is exactly the same as the operation of registering the voice index using the “reproduction start button” except that the operation is performed when information is recorded. Accordingly, the processing procedure is such that "reproduction" in the flowchart of FIG. 4 is replaced with "recording". The detailed description here is omitted. [Operation of Registering Audio Index in the Middle of Information] The “pause button” shown in FIG. 2 is also used for the operation of registering the audio index in the middle of a song, not at the beginning of the song. FIG.
As shown in (1), when the "pause button" is pressed (step S11), the operation is detected by the button operation detecting section 6, and the control section 14 determines whether or not the music is currently being reproduced (step S11). S12). At this time, no operation is performed unless the reproduction is being performed. However, if the reproduction is being performed, the control unit 14 instructs the information reproduction unit 11 to temporarily stop the reproduction operation (step S13). This is the normal operation when the “pause button” is pressed.
The control unit 14 simultaneously activates the audio detection unit 2 to start detection of audio in the input signal from the microphone 21 (step S14), and activates the timer 7 to activate the "pause button". The measurement of the pressed time is started (step S15). When the "pause button" that has been pressed is released (step S16), it is determined whether the time measured by the timer 7 at that time is equal to or longer than a predetermined threshold (for example, 0.5 seconds). (Step S17). Here, if the measurement time is less than the threshold value, no special operation is performed. That is, when the “pause button” is pressed for a short time, only a normal pause operation is performed.

On the other hand, if the measurement time is equal to or greater than the threshold, that is, if the “pause button” is continuously pressed for a predetermined time or more, the control unit 14 determines that the requested operation It is determined that the operation is a registration operation, and the following processing is performed. That is, it is determined whether or not the voice is detected by the voice detection unit 2 (step S18). If the voice is detected, the voice is converted into some parameters by the voice parameter conversion unit 3, and the voice is being reproduced. Is registered in the voice index storage memory 5 as a voice index indicating the current playback position (pause position) of the music piece (step S19). Since the voice index registered in this case is registered so as to indicate the middle of the music, for example, the voice index No. in the upper right diagram of FIG. As shown in FIG. 1, the song number, the midway address, and audio parameters P11, P12, P13,... Are registered. On the other hand, if no sound is detected in step S18, that is, if the "pause button" is continuously pressed for a predetermined time without inputting a sound, an address near the current reproduction position is determined. The registered voice index registered in is deleted (step S20). For example, the voice index No. in the upper right diagram of FIG. 1 is deleted.

[Information Search Operation 1] The "forward skip button" shown in FIG. 2 is also used as the start button for the information search operation. As shown in FIG. 6, when the "forward skip button" is pressed (step S21), the operation is detected by the button operation detection unit 6, and the voice detection unit 2 is activated by the control unit 14 and the microphone 21 Of the input signal is started (step S2).
2), the timer 7 is activated and the measurement of the time during which the “forward skip button” is pressed is started (step S23). When the "forward skip button" that has been pressed is released (step S24), the time measured by the timer 7 at that time is a predetermined threshold (for example, 0.5 seconds).
It is determined whether this is the case (step S25).
Here, when the measurement time is equal to or longer than the threshold, that is, when the “forward skip button” is continuously pressed for a predetermined time or longer, it is determined whether or not a sound is detected by the sound detection unit 2 ( Step S26) If a voice is detected, the voice is converted into several parameters by the voice parameter conversion unit 3, and the input voice parameter and the registered voice parameter registered in the voice index storage memory 5 are used. Is calculated by the above-described speech matching similarity calculation unit 4. As a similarity calculation method, DP matching or the like, which is a general technique in speech recognition, can be used. Then, from the registered voice parameters, one or a plurality of voice parameters having a similarity to the input voice parameter or more than a predetermined threshold are extracted, and the number of the voice index corresponding to the voice parameter is, for example, a high similarity. These are sequentially stored in the voice index search history memory 8 (see the lower right diagram in FIG. 3) (step S30). Next, a first-ranked voice index is selected from the voice indexes stored in the voice-index search history memory 8 (step S31), and the reproduction pointer is moved to the position of the selected voice index. Reproduction is started from the position (step S32).

On the other hand, if the time measured by the timer 7 is less than the threshold value in step S25, or if no voice is detected in step S26, the search history is stored in the voice index search history memory 8. Is determined (step S2).
7) If not stored, the playback pointer is moved to the beginning of the next song. That is, a normal operation when the “forward skip button” is pressed is performed. On the other hand, in step S27, the voice index search history memory 8
If it is determined that the search history is stored in the
The next-ranked voice index stored in the voice index search history memory 8 is selected (step S).
29) The reproduction pointer is moved to the position of the selected voice index, and reproduction is started from that position (step S32). [Information Retrieval Operation 2] The “backward skip button” shown in FIG. 2 is used to select the previous-ranked voice index stored in the voice index search history memory 8. As shown in FIG. 7, when the “backward skip button” is pressed (step S 41), the operation is detected by the button operation detection unit 6, and the search history is stored in the voice index search history memory 8 by the control unit 14. It is determined whether or not it has been performed (step S42). At this time, if the music is not stored, the reproduction pointer is moved to the head of the previous music piece. That is, a normal operation when the “backward skip button” is pressed is performed. On the other hand, in step S42, the voice index search history memory 8
If it is determined that the search history is stored in the
The previous voice index stored in the voice index search history memory 8 is selected (step S).
44), the reproduction pointer is moved to the position of the selected voice index, and reproduction is started from that position (step S45).

As described above, in the information storage / reproducing apparatus A1 according to the present embodiment, the speech matching similarity calculation unit 4 corresponds to the registered speech parameter whose similarity with the input speech parameter is equal to or greater than a predetermined threshold. One or more speech indexes to be extracted are extracted and ordered in the order of similarity, and a desired speech index can be selected based on the ordering. It is possible to easily detect a desired data position even in an environment in which the voice index decreases, or in a case where the same voice index is registered at a plurality of positions. Further, processing operations such as registration, deletion, and search of a voice index by voice are realized by continuously pressing operation buttons for normal operations such as reproduction and recording of information for a predetermined time or more. It is not necessary to provide an operation button dedicated to the processing by the above, so that the space can be effectively used even with a small device.

[0018]

In the above embodiment, the speech indexes extracted by the speech matching similarity calculator 4 are ordered in the order of the similarity of the speech parameters. For example, the speech indexes are ordered in the order of the storage positions indicated by the speech parameters. May be performed. The flowcharts shown in FIGS. 4 to 7 are examples, and the processing content assigned to each operation button can be set arbitrarily. Further, in the above embodiment, the information storage / reproducing apparatus for storing / reproducing audio information has been described as an example. However, the digital information to be handled is not limited to audio information, and various types of information such as still images, moving images, type information, etc. It is possible to target something. Further, the present invention is applicable to an apparatus that does not necessarily reproduce information but performs only information retrieval, for example.

[0019]

As described above, the present invention provides storage means for storing digital information, voice input means for inputting voice, and voice input by the voice input means.
An identification code storage means for storing the digital information stored in the storage means as an identification code indicating the position on the storage means; a voice input by the voice input means and an identification code stored in the identification code storage means; A search means for searching for the position of the digital information on the storage means on the basis of the code, the digital information storage and search device comprising: Similarity identification code ordering means for extracting one or more sounds similar to the voice inputted by the input means and performing a predetermined ordering, and identification code selection means for selecting a desired identification code in accordance with the ordering by the similarity identification code ordering means Since it is configured as a digital information storage and retrieval device characterized by having Even in an environment where the voice recognition accuracy is reduced due to such factors, or when the same voice index is registered at multiple positions, repetitive processing due to recognition errors can be prevented, and the desired data position can be easily detected. Becomes possible. Further, the operation of storing the identification code of the identification code storage means based on the voice input means, the instruction of the ordering operation by the similar identification code ordering means based on the voice input means, and the operation of the identification code selection means And the like can be realized by a predetermined operation on the storage / reproduction input means for inputting the instruction relating to the storage / reproduction of the digital information, so that input means such as dedicated buttons for processing by voice input are provided. Since there is no need to provide a space, it is possible to effectively utilize space even in a small device. If the processing by the voice input is performed by, for example, continuously operating the storage / reproduction input means for a certain period of time or more, it is easy to provide the storage / reproduction input means with a plurality of functions.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of an information storage / reproduction device A1 according to an embodiment of the present invention.

FIG. 2 is a schematic diagram showing operation buttons of the information storage / reproduction device A1.

FIG. 3 is an explanatory diagram showing a configuration example of information respectively stored in an information address management memory, a voice index storage memory, an information body storage memory, and a voice index search history memory in the information storage / reproduction device A1.

FIG. 4 is a flowchart showing a processing procedure of a voice index registration operation at the time of information reproduction using a “reproduction start button”.

FIG. 5 is a flowchart showing a processing procedure of an operation of registering a voice index in the middle of information using a “pause button”.

FIG. 6 is a flowchart showing a processing procedure of an information search operation using a “forward skip button”.

FIG. 7 is a flowchart illustrating a processing procedure of an information search operation using a “backward skip button”.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... A / D conversion part 2 ... Speech detection part 3 ... Speech parameter conversion part 4 ... Speech matching similarity calculation part (similar identification code ordering means) 5 ... Speech index storage memory (identification code storage means) 6 ... Button operation detection Unit 7: Timer 8: Voice index search history memory 9: Information address management memory 10: Information body storage memory (storage unit) 11: Information reproduction unit (reproduction unit) 12: D / A conversion unit 13: Information writing interface 14 ... Control unit (identification number selection means) 15 ... operation buttons (storage / playback input means) 21 ... microphone (voice input means)

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code FIG11B 27/10 C (72) Inventor Toshiaki Shimoda 1-5-5 Takatsukadai, Nishi-ku, Kobe-shi, Hyogo Kobe Steel, Ltd. Kobe General Co., Ltd. Within the Technical Research Institute (72) Inventor Yoshiro Nishimoto 1-5-5 Takatsukadai, Nishi-ku, Kobe City, Hyogo Prefecture Inside the Kobe Steel Research Institute, Kobe Steel Co., Ltd. No.2 Kobe Steel Ltd. Tokyo head office

Claims (7)

[Claims] 1. A storage means for storing digital information,
Voice input means for inputting voice, identification code storage means for storing the voice input by the voice input means as an identification code indicating the position of the digital information stored in the storage means on the storage means, A digital information storage and retrieval device comprising: retrieval means for retrieving the position of the digital information on the storage means based on the voice inputted by the voice input means and the identification code stored in the identification code storage means. A similar identification code ordering means for extracting one or a plurality of sounds similar to the voice input by the voice input means from the identification codes stored in the identification code storage means and performing a predetermined ordering; Identification code selecting means for selecting a desired identification code in accordance with the ordering by the similar identification code ordering means. Digital information storage and retrieval device for. 2. The digital information storage and retrieval device according to claim 1, further comprising a reproducing unit that reproduces the digital information retrieved by the retrieval unit. 3. The digital information storage and retrieval device according to claim 1, wherein the predetermined ordering is performed based on the similarity between each identification code and the voice input from the voice input means. 4. The digital information storage and retrieval apparatus according to claim 1, wherein the predetermined ordering is performed based on a storage position on the storage means indicated by each identification code. 5. An operation of storing said identification code in said identification code storage means based on said voice input means and an instruction of an ordering operation by said similar identification code ordering means based on said voice input means are stored in said digital information. 5. The digital information storage and retrieval device according to claim 2, wherein the digital information storage and retrieval device is realized by a predetermined operation on a storage / reproduction input means for inputting an instruction relating to reproduction. 6. The digital information storage / retrieval apparatus according to claim 5, wherein the instruction of the operation to the identification code selection means is realized by a predetermined operation on the storage / reproduction input means. 7. The digital information storage and retrieval device according to claim 5, wherein the predetermined operation is a continuous operation for a predetermined time or more.