JPH01255899A - Audio data editing device - Google Patents

Abstract

PURPOSE:To edit divided units having priority levels equal to or higher than an indicated priority level by storing audio data divided to editing units and adding priority levels to them and eliminating divided units having priority levels lower than the indicated priority level. CONSTITUTION:Audio data inputted from an audio input device 11 is subjected to the segment (editing unit) division processing by a segmenting circuit 13 and divided segments are stored in an audio data storage device 16. A priority level adding device 18 is used to add priority levels to respective segments, and a desired priority level is indicated by a priority level indicating device 19, and segments having priority levels lower than the indicated priority level are eliminated by a priority level selecting circuit 20, and they are outputted to an audio output device 17 if necessary. Thus, audio segments having priority levels equal to or higher than the indicated priority level are re-edited in the input order.

Description

[Detailed description of the invention] "Industrial application field" The present invention relates to an audio data editing device.

``Prior art'' Conventionally, there are devices that divide audio data into sentence units by setting a threshold value for silent time during input, but there is a function to add priority to the division units and the priority level. There is no device that has the function of selecting audio data using .

"Means for Solving the Problem" According to the present invention, audio data divided into editing units (segments) is stored in the audio data storage device 1, and priority is added to the audio data divided into editing units. Attached by device. When the priority is indicated by the priority indicating device,
A division unit (segment) having a priority lower than the designated priority is removed by the priority selection circuit, and a division unit having a priority higher than the designated priority is edited.

Further, the time length of each division unit (segment) is measured by a segment time length measuring circuit, and segment time length information is added to the segment. When the edited time length is specified by the message time length specifying device, the division units with low priority are sequentially removed by the priority selection circuit so that the time length becomes a value close to the specified time length, and the division units with high priority are removed. (segment)
A message of the specified length consisting of the following messages is obtained.

Embodiment FIG. 1 shows an embodiment of the present invention, in which a voice input device 11. Silent time measurement circuit 12, segmentation circuit 13, segment start time information addition circuit 14, segment address addition circuit 15, audio data storage device 16,
It consists of an audio output device 17, a priority addition device 18, a priority instruction device 19, a priority selection circuit 20, a message duration designation device 21, and a segment duration measurement circuit 22.

First, the silent time measuring circuit 12 measures the silent time between the voice data inputted from the voice input device 11 using the silent time measuring circuit 12 . Next, a threshold is set for the silent time of the audio, and when the silent time exceeds the threshold "t", the segmentation circuit 13 performs segment (editing unit) division processing. The segment start time information addition circuit 1-4 and the segment address addition circuit 15 automatically add the start time of the segment and the file address of the segment as management information to each segment that has been divided into segments through this process. The silent time is compressed and stored in the audio data storage device 16.

When adding priorities, the operator uses the priority adding device 18 to assign priorities to each segment while listening to the audio data output in segments from the audio output device 17 according to the semantic importance of each segment. Add degree. When selectively outputting segments based on priority, the priority instruction device 19 instructs the desired priority, the priority selection circuit 20 removes audio segments with less than the specified priority, and outputs the audio segment as necessary. Output device 17
Output to.

In addition, when shortening a voice message that exceeds a desired time to a desired time, the message time length designation circuit 21 designates the desired time length, and the segment time length measurement circuit 21
After measuring the time length of each segment based on the number of constituent data of the segment in step 2, the priority selection circuit 20 sequentially removes segments starting from the lowest priority so that the total time length of the segments is close to the desired time length. Perform long edits.

FIG. 2 is an example of management information for managing the n-th segment,
It consists of a segment number 23, priority 24, segment start time information 25, and segment address 26.

FIG. 3(a) is an example of an audio segment management information file after the segment priority is assigned in addition to the segment number, segment start time information, and segment address. For example, when the priority is specified as 11311. In this case, the file is changed as shown in Figure 3(b), and only segments with a priority level of 3 or higher are added at a certain time interval (for example, at the time of input) between the end time of the previous segment and the start time of the next segment. The silent time threshold "t") is set and the data is re-edited in the input order.

FIG. 4(a) is an example of a management information file that has been segmented at the time of input and has been given a priority. Each segment has the time length shown in the figure, and the total time length of the segments is 38 (see , :l. Also, assume that the total time length of the no-data section is 20[5eel.This total time length is 60
If you shorten the message [□□□] to a message with a total time of 3% [%], considering that it includes 10 [(8)] of silent time, the total time length of the segment will be reduced to 20 [sec].
]. Therefore, by measuring the total time length in order from the highest priority segments and removing data with priority less than "3" so that the total time length is close to 20, as shown in Figure 4 (b). The file is changed, the total time length of the segments becomes 18 [(8)], and the message is shortened by removing segments of low importance.

"Effects of the Invention" According to the present invention, audio segments with designated priorities a to h can be re-edited in the order of input, so audio segments that do not need to be output, such as meaningless exclamations, are not output depending on the priority designation. In addition, voice messages that exceed a desired time can be kept within the time without omitting important parts. By utilizing this invention, it becomes possible to create audio minutes, etc. according to priority based on one audio file.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing an example of segment management information, and FIG. 3(a) is a diagram showing an example of a segment management information file before priority is specified.
FIG. 3(b) is a diagram showing an example of the segment management information file after the eel priority is specified, FIG. 4(a) is a diagram showing an example of the segment management information file before the priority is specified, and FIG. FIG. 7 is a diagram showing an example of a segment management information file after message shortening. Patent distributor: Nippon Telegraph and Telephone Corporation Representative: Takuki Kusano 1
Figure (Q) Total 38; Z (b) 20 [sec] fingers

Claims (2)

[Claims] (1) A means for storing audio data divided into editing units, a means for adding a priority to the division unit, a means for instructing the priority, and a division having a priority lower than the designated priority. An audio data editing device comprising: means for removing units. (2) means for storing audio data divided into editing units; means for adding priority to the division units; means for measuring the time length of each division unit; and means for adding segment time length information; An audio data editing device comprising: means for specifying a later time length; and means for sequentially removing division units with low priority so that the time length becomes a value close to the specified time length.