JPH10283162A - Retrieval system for voice information file - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a retrieval system for a voice information file which does not need a display device and also fast retrieves desired voice information even without depending on a number. SOLUTION: A voice information file retrieval system which retrieves desired voice information file among plural voice information files is provided with an index area 2c which stores summary information that summarizes contents of each voice information file 2d and the keyword of the file 2d as index information, a controlling part 1 and a voice reproducing part 7 which are means that successively reproduce the index information of each file 2d stored in the area 2c as a voice at the time of reproducing voice information, specifies one piece of index information from plural index voice information which are reproduced by the parts 1 and 7, etc., and reproduces the voice information of a voice information index that is associated with the specified index information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information file search technique for reproducing audio information stored in a storage medium, and more particularly to a search method for an audio information file capable of performing a high-speed search without using a display device. Things.

[0002]

2. Description of the Related Art Conventionally, when reproducing desired audio information from a plurality of audio information stored in a storage medium such as a magnetic tape, generally, the storage medium is fast-forwarded or rewinded. Thus, a method has been adopted in which voice information is reproduced each time and desired voice information is searched for by trial and error. However, since such a search method is inefficient, in recent years, a number has been added to voice information and a search based on the added number has been performed, or a keyword has been added as in a multimedia system. Search methods are provided as well as file systems.
Further, in the document reading-out apparatus disclosed in Japanese Patent Application Laid-Open No. 5-281987, contents are extracted from the heading of the document information composed of character code strings, high-speed search is performed, and speech is synthesized from the extracted character code strings. Output. In the information reproducing apparatus disclosed in Japanese Patent Application Laid-Open No. 6-325587, high-speed search is performed by sequentially reproducing only the head of each audio information. Further, in the information providing apparatus for the visually impaired disclosed in Japanese Patent Application Laid-Open No. Hei 7-36377, a menu can be output by voice in a menu-type information search using a display device.

[0003]

However, in the above-mentioned prior art, the method of searching by number requires that the user knows what the audio information indicated by each number is. In a search method similar to the multimedia system and a method disclosed in Japanese Patent Application Laid-Open No. Hei 7-36377, a display device is required. Also, Japanese Patent Application Laid-Open No. 5-2819
JP-A-6-3
The search based on the head portion of the audio information disclosed in Japanese Patent No. 25587 has a problem that it is difficult to determine whether the audio information is the desired audio information from the headline or the head portion. The problem to be solved by the present invention is to solve the above-mentioned drawbacks of the conventional technology, and to provide a voice information file search method that can search for desired voice information at high speed without requiring a display device and without relying on a number. Is to provide.

[0004]

According to the first aspect of the present invention, there is provided a voice information file search method for searching for a desired voice information file from a plurality of voice information files. Index storage means for storing summary information summarizing the content of each voice information file or keywords of the voice information file as index information;
Index reproducing means for sequentially reproducing, as audio, index information of each audio information file stored in the index storing means, and identifying one index information from a plurality of index audio information reproduced by the index reproducing means. By reproducing the audio information of the audio information file associated with the specified index information, the user can listen to a plurality of sequentially reproduced index audio information such as keywords and summary information, thereby obtaining desired audio information. Added support for playback. According to a second aspect of the present invention, in a voice information file search method for searching for a desired voice information file from a plurality of voice information files, summary information summarizing the contents of each voice information file or the voice information file An index storage unit for storing the keyword of the file as index information composed of a character code string; and, when reproducing the audio information, when a character code string is designated, the index information including the designated character code string is stored. A short character string related to desired voice information from a keyboard or the like by comprising an index search means for searching and reproducing the voice information of the voice information file based on the search result by the index search means. Thus, the desired audio information can be reproduced. Further, the invention according to claim 3 further includes a mark adding means for adding mark information to index information or index management information based on the apparatus configuration according to claim 1 or 2, By configuring the index information to be searched using the addition of the mark information as a search condition, it is possible to quickly reproduce the audio information by adding a mark to the index information of the desired audio information. I made it.

According to the invention described in claim 4, according to claim 1,
Or mark addition means for adding mark information accompanied by mark position information to index information or index management information on the premise of the device configuration described in or 2,
At the time of reproducing the audio information, the audio information is reproduced from the position indicated by the mark position information in the audio information file associated with the index information or the index management information. Is added so that a desired portion of the audio information can be quickly reproduced. According to a fifth aspect of the present invention, based on the device configuration according to any one of the first to fourth aspects, management information is stored in association with each audio information file. As one of the management information, by providing an erasure mark adding means for writing an erasure mark for prohibiting the subsequent access to the audio information file, the audio information with the erasure mark can be excluded from the search target.

[0006]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram showing an example of an embodiment of a recording / reproducing apparatus to which a voice information file search method according to the present invention is applied. As shown in the figure, the recording / reproducing apparatus of this embodiment includes a control unit 1 for performing recording / reproducing control and file management, a plurality of audio information files 2d (2d1, 2d2,...), And connection to each file ( A storage medium 2 having an index area 2c (2c1, 2c2,...) In which the associated index information is stored; an operation unit 3 for performing operations and text input; a microphone 4 for inputting voice; An encoding unit 5 for digitizing and encoding the analog audio signal;
It includes a speaker 6 for outputting voice, a voice reproducing unit 7 for decoding coded voice information to convert it into an analog signal, and synthesizing voice from text information, and a RAM 8 for temporarily storing management information and voice information. ing. The storage medium 2 is a magnetic tape, a writable CD, or the like. Further, the file management area 2a, the index management area 2b, the index area 2c, etc. may be stored in a hard disk or a RAM backed up by a battery.

Next, the operation of the recording / reproducing apparatus according to the present embodiment will be described. FIG. 2 is a flowchart showing a processing procedure at the time of recording by the recording / reproducing apparatus of the present embodiment. As shown in the figure, at the time of recording, first, a user presses a recording key in the operation unit 3 (S1), and inputs a file name from the operation unit 3 if necessary. When the control unit 1 recognizes that the recording key has been pressed, it obtains the last address information of the file from the file management area 2a as shown in FIG.
The final address information when no audio information file is stored in the storage medium 2 is a predetermined address set in advance. When the control unit 1 obtains the last address information, the index management area 2b (see FIG. 4) and the index area 2c of the audio information to be stored from now on
(See FIG. 5), the respective addresses at the head of the audio information file are obtained (in this embodiment, the index area 2c has a fixed length). Then, when the file name is input, the obtained file name is written in the file name area in the index management area 2b. Subsequently, the user inputs the index information (S2). The index information is, for example, a keyword or a summary of audio information to be stored (recorded) from now on, and is input from the operation unit 3 as text (character string) or from the microphone 4 as audio information. In the case of text, the control unit 1 acquires the text character code string input from the operation unit 3 and stores the text character code string as it is in the index area 2c. Is converted to a phonetic symbol string and stored,
In the case of audio information, the audio information is digitized by the encoding unit 5, and further encoded and stored by PCM or ADPCM (S3) (see FIG. 5).

[0008] When the input of the index information is completed, the user performs a key input to that effect, and starts the input of the voice stored in the voice information file from the microphone 4 (S).
4). Then, the analog signal from the microphone 4 is digitized in the encoding unit 5, further encoded by the PCM or ADPCM method, and passed to the control unit 1 via the RAM 8. In this way, the control unit 1 stores the acquired audio information as the audio information file 2d (S
5). The storage address has already been determined as described above. The information stored as the audio information file 2d may be text information as in the case of the index information. In this case, it is input from the operation unit 3 and stored in the same manner as the index information (see FIG. 5).

At the time of the above-mentioned recording (at the time of storage), the control unit 1 acquires a storage area for the audio information file 2d in units of one block at any time and counts the number of acquired blocks. When the storage is completed, the file management information in the file management area 2a is updated (S6). That is, the last address of the audio information file 2d is calculated according to the acquired number of blocks, the content of the last address information area (see FIG. 3) of the file is updated, and the content of the remaining capacity information area is updated. Further, the control unit 1 calculates the file size (text file size) of the audio information file according to the acquired number of blocks, and writes the file size in the text file size area in the index management area 2b (S7) (see FIG. 4). Further, the next address of the last address of the audio information file is written in the subsequent index pointer area in the index management area 2b, and the index (the previous index pointer area which is the head address of the next index management area 2b) is stored in this address. Management area 2
Write the start address of b.

As shown in FIG. 4, in the recording / reproducing apparatus of this embodiment, a marker can be inserted in the audio information in the audio information file 2d. That is, when the mark key in the operation unit 3 is pressed during the input of the voice information, the control unit 1 writes the voice information storage address at that time in the text marker position information area in the index management area 2b.
In the example shown in FIG. 4, a plurality of marker positions can be set. Thus, in this embodiment, sound reproduction from a plurality of arbitrary marker positions is possible.
Also, the mark information can be used for audio information file search. In this case, one mark information setting area is provided in the index management area 2b, and as described later, only the index information in which the marker is set can be reproduced.

As shown in FIG. 3, an erase mark / bit area is provided in the file management area 2a,
Access to the audio information file to which the deletion mark has been added can be prohibited. That is, at the time of retrieval, the audio information file to which the erasure mark has been added is treated as if it does not exist. In addition, a configuration in which the erasure mark is set in the index management area 2b is also possible. In FIG. 3, the voice utterance environment information area includes voice data types (male sounds, female sounds, robot sounds, and the like) and Japanese conversion conditions (for example, information for determining whether or not to use numbers as stick readings). When voice information is text information, such as reading speed, voice pitch, and accent strength, information necessary for voice synthesis is set in advance.

The file sum check area stores the checksum of the audio information file stored so far. And when writing occurs,
While the writing is being performed, a process of adding the written amount to the checksum is executed. When the writing is completed, the audio information in all the stored audio information files 2d is read, and the checksum is calculated. Is compared with the checksum in the file sum check area. In this way, not only is it checked whether or not the data has been correctly written, but also whether or not audio data written after the writing has been corrupted.

The voice utterance condition area shown in FIG. 4 is used to output voice information of the connected voice information file under a condition different from the voice utterance environment information in the file management area 2a. The voice utterance conditions of different parts are stored.

FIG. 6 is a flowchart showing a processing procedure at the time of reproduction by the recording / reproducing apparatus of the present embodiment. As shown in the figure, at the time of reproduction, the user instructs a search using a search key or the like (S11). Then, the control unit 1 reads out each management information in the file management area 2a into the RAM 8 and refers to the erasure mark bit area. If the erasure mark is not added to the first audio information file, the first index management area 2b1 is read. The contents are read out to the RAM 8. further,
The control section 1 reads the index information of the first index area 2c1 from the storage medium 2. Then, in the case of voice output by the voice synthesis method, the voice reproduction environment information in the file management area or the voice utterance condition in the index management area is given to the voice reproduction unit 7. In this way, the voice reproducing unit 7 synthesizes a voice signal from the index information according to the voice utterance environment information or the voice utterance condition, and outputs a voice to the speaker 6 (S12). The content of the voice is the summary information of the voice information file 2d1 linked to the index information or the keyword of the voice information. In the above, when the audio information is information encoded by PCM or ADPCM,
The audio reproduction unit 7 performs decoding and conversion into an analog signal.

The user listens to the output index information, determines whether or not to output the text (audio information in the audio information file), and presses the playback key in the operation unit 3 to output the text. That is, the control unit 1 monitors the pressing of the reproduction key while outputting the index sound (S13). If the index information being output is output to the end without the pressing of the reproduction key (No in S13). The contents of the next index management area 2b2 are called into the RAM 8 in accordance with the subsequent index pointer in the index management area (S14). Subsequently, the control unit 1
It is determined whether or not the audio information file 2d2 has an erasure mark by referring to the erasure mark bit of the file management information in the AM 8 (S15). If the erasure mark has been added (Yes in S15), furthermore, The contents of the next index management area 2b3 are called (S14).

On the other hand, if it is determined in step S15 that the erasure mark has not been added (S15)
No), the index information concatenated as described above is output as voice (S12). When the pressing of the reproduction key is detected while a plurality of index information are sequentially output as audio (Yes in S13), the control unit 1 then stores the RAM in the RAM.
Then, it is determined whether or not the marker position information is added to the index information called in step 8 (S16).
If it is determined that the audio information is added (Yes in S16), the reading of the audio information is started from the marker position (address) in the corresponding audio information file 2d according to the position information (S17), and the index information is read. Is output to the speaker 6 in the same manner as in the case of the output (S1).
8). This audio output is stopped when the user presses the stop key in the operation unit 3 or the audio information in the audio information file is output to the end. However, when the stop is performed by pressing the stop key, if there is a next marker (Yes in S19), the sound is output again from the next marker position by pressing the start key or the like (S19).
17, S18).

If it is determined in step S16 that a marker has not been added (N in S16,
o), audio is output from the beginning of the linked audio information file 2d (S18), and when the user presses the stop key in the operation unit 3 or the audio information in the audio information file is output to the end. Stop. In a configuration in which only one marker is added to the index management information and this is used as a condition for file search, when a search instruction is given when sequentially outputting index information (S1).
1) The control unit 1 reads out the contents of the index management area 2b one after another, determines whether or not a marker is added, and outputs the linked index information as audio only when the marker is added. If the index information is stored as text information (character code string), instead of sequentially outputting the index information from the beginning, a character code string is designated by the operation unit 3 and the character code string is included in the index information. It is also possible to sequentially output only the index information. That is, when acquiring the designated character code string, the control unit 1 sequentially reads the information in the index area 2c from the beginning, and determines whether or not the designated character code string is included in the information. If it is included, the operation of outputting the index information is repeated. Note that the above-described audio information file search method of the present invention employs a recording / reproducing apparatus using various storage media such as a magnetic tape, a flexible disk, a hard disk, a writable CD (CD-R, CD-E, etc.) or other It can be applied to the device of the above.

[0018]

As described above, according to the present invention, the following excellent effects can be exhibited. According to the first aspect of the present invention, desired audio information can be reproduced by listening to a plurality of sequentially reproduced index audio information such as keywords and summary information, so that a display device is not required and the number does not depend on the number. Desired voice information can be searched at high speed.
According to the second aspect of the present invention, the desired voice information can be reproduced by inputting a short character string related to the desired voice information from a keyboard or the like, so that the desired voice information can be searched at a higher speed. According to the third aspect of the present invention, the mark is added to the index information of the desired audio information, so that the reproduction time of the index information is shortened, so that the audio information can be reproduced quickly. According to the fourth aspect of the present invention, by adding a mark to a desired position of the audio information, the reproduction can be started from the mark position, so that the desired portion of the audio information can be quickly reproduced. Claim 5
According to the invention described in (1), audio information with an erasure mark can be excluded from the search target, so that the search time can be reduced, and after that, the erasure mark can be removed and returned to the original state, and the audio information file can be deleted. It can be easily deleted.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram showing an example of an embodiment of a recording / playback apparatus to which a search method according to the present invention is applied.

FIG. 2 is a flowchart showing a processing procedure when recording by the recording / reproducing apparatus according to the embodiment of the present invention.

FIG. 3 is a file configuration diagram of a main part of the recording / reproducing apparatus according to the embodiment of the present invention.

FIG. 4 is a data configuration diagram of a main part of the recording / reproducing apparatus according to the embodiment of the present invention.

FIG. 5 is another data configuration diagram of a main part of the recording / reproducing apparatus according to the embodiment of the present invention.

FIG. 6 is a flowchart showing a processing procedure at the time of reproduction of the recording / reproducing apparatus according to the embodiment of the present invention.

[Explanation of symbols]

1 control unit (index reproduction means, index search means), 2 storage medium (index storage means), 3
Operation unit (mark information adding unit, erasing mark information adding unit), 4 microphone, 5 encoding unit, 6 speaker, 7
Audio playback unit (index playback means), 8 RAM

Claims (5)

[Claims] In a voice information file search method for searching for a desired voice information file from a plurality of voice information files, summary information summarizing the contents of each voice information file or a keyword of the voice information file is indexed. An index storage means for storing information as information, and an index reproduction means for sequentially reproducing, as audio, index information of each audio information file stored in the index storage means when reproducing the audio information, A plurality of indexed audio information reproduced by the method, one index information is specified, and the audio information of the audio information file associated with the specified index information is configured to be reproduced. Search method. 2. A speech information file search method for retrieving a desired speech information file from a plurality of speech information files, wherein summary information summarizing the contents of each speech information file or a keyword of the speech information file is written in characters. Index storage means for storing index information composed of code strings; and index search means for retrieving index information including the specified character code string when reproducing a sound information, when a character code string is specified. Wherein the audio information of the audio information file is reproduced based on the search result by the index search means. 3. The audio information file search method according to claim 1, further comprising a mark adding means for adding mark information to index information or index management information, wherein said mark information is added when audio information is reproduced. A search method for a voice information file, characterized in that index information is searched using a search condition as a search condition. 4. The audio information file search method according to claim 1, further comprising: mark addition means for adding mark information with mark position information to index information or index management information;
An audio information file search method, wherein audio information is reproduced from a position indicated by the mark position information in an audio information file associated with the index information or the index management information. 5. The voice information file search method according to claim 1, wherein management information is stored in association with each voice information file, and one of said management information is stored. And a deletion mark adding means for writing a deletion mark for prohibiting subsequent access to the voice information file.