JPH0442680B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an audio editing device used for audio storage services and the like.

(Prior Art and its Problems) Conventionally, in audio storage services, processing such as compression of the amount of data through encoding has been performed, but no processing has been performed on the content of the audio. Therefore, it has been impossible to insert pause sections as needed to match the writing speed when writing down the output audio. For example, in a paper published in IEICE Technical Report SE79-98 entitled "Sound Storage Service Concept," it mentions data compression, storage, and playback through audio encoding; No in-depth processing has been done on the contents.

(Object of the Invention) An object of the present invention is to provide a speech editing device that inserts a pause section for each accent phrase to facilitate dictation by listening by detecting accent phrase boundaries in input speech. be.

(Means for Solving the Problems) In order to solve the above-mentioned problems, the audio editing device provided by the present invention includes pitch extraction means for extracting a time series of pitch frequencies of input audio; means for detecting addresses of accent phrase boundaries of the speech from a time series of the speech, means for outputting the speech to the waveform memory for each of the accent phrases separated by the addresses of the accent phrase boundaries, and outputting from the waveform memory. and a boundary processing means for inserting a pause section into the voice at each accent phrase boundary.

(Principle of the invention) The principle of the invention will now be explained. The audio editing device according to the present invention estimates the accent phrase boundary from the interval where the pitch frequency is a valley, and when outputting it, inserts a pause interval at the accent phrase boundary, thereby facilitating dictation by listening. It is. Hereinafter, the principle of the present invention will be explained based on an example. FIG. 2 is an example showing the change in pitch frequency over time for the utterance "I have been losing money ever since I was a child." The vertical axis is the pitch frequency of the input audio, and the horizontal axis is the time axis. This time series of pitch frequencies is described, for example, in a paper published in the Proceedings of the Acoustical Society of Japan (October 1977), pages 35-36, entitled ``A method for optimal selection of pitch periodic sequences.'' It can be obtained by the following method. The part where the locus of the pitch frequency forms a gentle valley is detected as an accent phrase boundary. A gentle valley is, for example, a pitch time series p(i), i=1...
…N moving average P(i)=p(i-1)+2 _p (i)+p(i+1)/4
It can be obtained as the minimum part of . Points a and b in FIG. 2 are examples of accent phrase boundaries. When outputting the accumulated audio, by inserting pauses as appropriate at accent phrase boundaries, it is possible to output the audio at a speed suitable for the user to listen to and write down.

(Example) Next, an example of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing one embodiment of the present invention. Input audio 1 is held in waveform memory 5,
It is also supplied to the pitch extraction section 2. The pitch extractor 2 calculates a time series of pitch frequencies from the supplied input speech 1 and supplies it to the accent phrase boundary detector 3. The accent phrase boundary detection unit 3 detects a valley portion from the time series of pitch frequencies supplied, and outputs the address on the waveform of that portion to the boundary address memory 4. Upon input of the output command 8, the control circuit 6 uses the phrase boundary address supplied from the boundary address memory 4 to output the speech waveform from the waveform memory to the boundary processing unit 7 for each accented phrase. The boundary processing unit 7 adds a pause section to the supplied speech waveform of each accent phrase and outputs the speech waveform.

(Effects of the Invention) According to the present invention, it is possible to obtain a speech editing device that inserts a pause section for each accent phrase to facilitate dictation by listening by detecting accent phrase boundaries in input speech.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention;
FIG. 2 is a pitch frequency locus diagram shown to explain the principle of the present invention. 2... Pitch extraction section, 3... Accent phrase boundary detection section, 4... Boundary address memory, 5... Waveform memory, 6... Control circuit, 7... Boundary processing section.

Claims (1)

[Claims] 1. Pitch extraction means for extracting a time series of pitch frequencies of input speech, means for detecting addresses of boundaries of accent phrases of the voice from the time series of pitch frequencies, and means for detecting addresses of boundaries of accent phrases of the voice, separated by addresses of boundaries of accent phrases. A voice editing device comprising means for outputting the voice to the waveform memory for each accent phrase, and boundary processing means for inserting a pause section at each accent phrase boundary into the voice output from the waveform memory.