JPH04208758A - Image filing device - Google Patents

Abstract

PURPOSE:To facilitate the registration of images from a remote place by providing a facsimile signal receiving means, a means for recognizing the voices from circuit and a means for recording the result of voice recognition as code data in the image filing device. CONSTITUTION:A voice synthesizing section 29 outputs a voice synthesis signal to a circuit output section 36. A sound synthesizing section 20 and a parameter extracting section 21 starts the analysis of input voices and the extraction of parameters. A voice section detecting section 22 detects voice sections from the results thereof. A phoneme segmentation section 23 detects phoneme boundaries and identifies the phonemes of the phoneme sections. A word matching section 28 obtains the results of word recognition. The results of the recognition are recorded in the form of the code data as document retrieval keywords. After all the necessary keywords are inputted, facsimile transmission is requested by the synthesized sounds from the voice synthesizing section 29. A facsimile receiving section 30, thereafter, receives the facsimile signals from the circuits and an image decoding section 31 stores the received signals into an image memory 32.

Description

[Detailed description of the invention]

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image filing apparatus; this invention relates to an image filing apparatus using a large capacity recording medium such as an optical disk. [0002]

2. Description of the Related Art An example of the configuration of a conventional image filing device is shown in FIG. In this figure, reference numeral 1 is a main CPU that controls the operation of the entire image filing device;
3 is a hard disk control unit that controls writing and reading operations to a hard disk that is a recording medium; 3 is an optical disk control unit that controls writing and reading operations to and from an optical disk that is a recording medium; 4 is an image memory for storing image data; 5 compresses image data,
This is an image converter that decompresses. Further, reference numeral 6 denotes a scanner control unit for controlling image input operation;
Reference numeral 8 designates a printer control unit that controls printing operations, 8 a frame memory in which image data is developed, and 9 a facsimile communication control unit that controls various operations necessary for facsimile communication. Further, reference numeral 10 is a system bus connected between the main CPUI and each operation section and through which various command data and management data are sent; 11 is an image bus through which image data is sent and received between the main CPUI and various operation sections; 12 13 is a hard disk in which data is stored; 13 is an optical disk controlled by the optical disk controller 3; 14 is an image scanner for inputting images; 15 is a printer for printing images; and 16 is a display device for displaying images.
RT, 17 is a public line through which data for facsimile communication is sent, and 18 is a keyboard through which various commands are input to the main CPUI. [0003] In such an image filing device, an image to be registered is read by an image scanner 14 and then stored in an image memory 4. The image is then compressed and encoded by the image converter 5 and recorded on the optical disc 13 under the control of the optical disc controller 3. On the other hand, keywords and document names necessary to search for this image are input from the keyboard 18 and recorded on the hard disk 12 or optical disk 13. This optical disc 1
3, the main CPU searches for the address of the desired image on the optical disk 13 from the files that make up the database on the hard disk 12 based on the keywords and document names entered from the keyboard 18. search for. Then, the image data corresponding to this address data is transferred to the optical disk controller 1 by the roll unit 3.
Under the control of C
Display on RT16. [0004]

[Problems to be Solved by the Invention] However, in such a conventional image filing device, since the image scanner 14 is used as an image input means, it is difficult to access the image from a remote location without using some kind of communication means. could not be registered. Therefore, a facsimile device or the like is used as a communication means, but when communicating with this device, the image data is first received by the facsimile receiving device, and then the image is transferred to the image scanner 1.
4, and search keywords and document names had to be entered from the keyboard. [00051] The present invention has been made in view of the above-mentioned problems, and an object thereof is to provide an image filing device that can easily register images from a remote location. [0006]

[Means for Solving the Problems] In order to achieve the above object, the present invention includes a facsimile receiving means, a means for recognizing a voice signal from a line, and a means for recording a voice recognition result as code data in an image filing device. The gist is that it has the means. [0007]

[Operation] With the above configuration, when registering a document from a remote location, the image data is automatically recorded on the optical disk by the facsimile receiving means, and search information such as keywords and document numbers can be retrieved in advance by voice recognition means such as a telephone. By entering this, you can separate the operations on the sending and receiving sides during registration. Therefore, no operation is required on the receiving side at the time of registration, and registration can be easily performed from a remote location. [0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of an image filing device according to the present invention. In this figure, reference numeral 20 is an acoustic analysis section for extracting speech recognition parameters, and for this acoustic analysis section 20, for example, a filter bank or an LPG analysis device is used. Reference numeral 21 denotes a speech recognition parameter extraction unit, which has a function of extracting various parameters such as a power component, a vowel component, a consonant component, a voiced component, and an unvoiced component. 22 is a voice section detection unit;
It has a function to detect the beginning and end of a word. 23 is a phoneme segmentation unit that detects phoneme boundaries from the extracted parameters. Reference numeral 24 denotes a phoneme standard pattern storage unit in which phoneme standard patterns expressed by various parameters are stored;
25 is a phoneme identification unit that identifies phonemes in each phoneme segment based on the degree of similarity with the standard pattern stored in the standard pattern storage unit 24. Reference numeral 26 is a word dictionary expressed in phoneme symbols, 27 is a conflation matrix showing what kind of phoneme each phoneme in the word dictionary is likely to be identified with, and 28 is a matching between each identified phoneme and the word dictionary. This is a word matching section that takes Further, reference numeral 29 is a voice synthesis unit for prompting a registered operator at a remote location to input the next input;
30 is a facsimile receiving section, 31 is an image decoding section that decodes the received signal into image data, and 32 is an image memory that records the decoded image data. 33 is a data management unit for recording voice recognition results as keywords, and 34 is an optical disk device for recording image data and keywords. 35 is a line input section for inputting signals from the public line, and 36 is a line output section for outputting signals to the public line. [0009] The operation of the image filing device having such a configuration will be explained. FIG. 2 is a flowchart showing an example of the document registration operation. In this figure, first, when the line input section 35 is connected (step 41), the speech synthesis section 29 outputs a speech synthesis signal to the line output section 36 in order to prompt the other party to generate a keyword (step 4).
2). Next, the acoustic analysis section 20 and the parameter extraction section 21 immediately start analyzing the input speech and extracting parameters (step 43), and based on the results, the speech section detection section 22
The phoneme segmentation unit 23 detects a phoneme interval (step 44), and the phoneme segmentation unit 23 detects a phoneme boundary and identifies the phoneme in the phoneme interval (step 45). Next, word matching section 2
Step 8 calculates the degree of similarity between the identified phoneme sequence and the word dictionary 26 in consideration of the confusion matrix 27 to obtain a word recognition result (step 46). This recognition result is transferred from the word matching unit 28 to the database management unit 33 and recorded as a document search keyword in the form of code data (step 47). The word matching unit 28 determines whether the document search keyword recorded in the database management unit 33 is the last of the required keyword group (Step 48), and if necessary, instructs the speech synthesis unit 29 to select the next keyword. The generation of is prompted by a synthesized sound (step 49). After all the necessary keywords have been input in this way, a request for facsimile transmission is made using synthesized speech from the speech synthesis section 29 (step 50).
). Thereafter, the facsimile receiving section 30 immediately enters a receiving state and receives a facsimile signal from the line, and the image decoding section 31 decodes the received signal as image data and stores it in the image memory 32 (step 51). Finally, the image data on the image memory 32 and the data in the database management section 33 are transferred to the optical disk 34.
Step 52), the registration sequence ends. [00101] Effects of the Invention As described above, according to the present invention, image data is automatically recorded and registered on an optical disc by the facsimile receiving means. Therefore, there is no need to read the image using the image scanner and input the image data again. Further, the document name and search keyword information can be registered as code data by voice recognition means, and operations on the sending side and operations on the receiving side can be separated. Therefore, it is possible to eliminate the need for operation on the receiving side at the time of registration, and it is possible to simplify registration from a remote location.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of an image filing device according to the present invention.

FIG. 2 is a flowchart showing an example of document registration operation in the embodiment.

[Fig. 3] Block diagram showing a conventional example of an image filing device

[Explanation of symbols]

20 Acoustic analysis section 21 Parameter extraction section 22 Speech interval detection section 23 Phoneme segmentation section 24 Phoneme standard pattern storage section 25 Phoneme identification section 26 Word dictionary 27 Confusion matrix 28 Word matching section 29 Speech synthesis section 30 Facsimile transmission section 31 Image decoding Section 32 Image memory 33 Database management section 34 Optical disk device 35 Line input section 36 Line output section

[Figure 1]

[Figure 3]

Claims (3)

[Claims] 1. A facsimile receiving means for receiving image data, a recording medium for storing data received by the facsimile receiving means, a means for inputting a voice signal from a circuit, a voice recognition means, and a voice recognition device. An image filing device having means for recording results as code data. 2. The image filing device according to claim 1, wherein the information on the voice recognition result is information related to a search keyword for the image data. 3. The image filing device according to claim 1, wherein the recording medium is an optical disc.