JPS6155698A - Recorder/reproducer - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Technical Field] The present invention provides a recording/playback device 1c having a fast-forward audio playback function.
It is related to

[Background technology]

Currently, there are recording and playback machines on the market that record audio and play it back in fast forward. This is to reproduce the recorded original sound of the waveform shown in FIG. 3(a) VC at twice the speed of recording, and if left as is, the frequency will be doubled as shown in FIG. 3(b). In order to make this the same frequency as the original sound frequency, K performs time expansion as shown in FIG. Then, the waveform in the dotted line portion is left over, so by performing a thinning reproduction process to discard this dotted line portion, the sound is reproduced at the same frequency as the original sound even during double speed reproduction. However, when the speaking speed is changed, the length of non-stationary parts such as consonants does not change much.
Considering the phenomenon in which the length of a steady part such as a vowel changes, it is not determined whether the part to be thinned out and reproduced is a steady part or an unsteady part, so the reproducibility is considered to be not very good, and the disadvantage is that the sound quality is not very good. have.

[Purpose of the invention]

The present invention has been made in view of the above-mentioned problems, and its purpose is to provide a recording/playback device that can play back recorded audio in fast forward motion while suppressing deterioration of sound quality as much as possible.

[Disclosure of the invention]

FIG. 1 shows an embodiment of the present invention, and (1) shows an A/D
A converter that converts audio input into a digital signal. (
2) is a speech analysis circuit that extracts synthesized speech parameters from the output of the A/D converter (1). (3) is a D/A converter, (4) is a chip recorder as a memory circuit, and (6)
) is an A/D converter. In this embodiment, since the target is a long meeting such as a meeting, the amount of data is large. Therefore, in order to be able to store a large amount of data at low cost as a memory circuit, the synthesized sound parameter (here nth is called (Pn)), which is the output of the speech analysis circuit (2), is A digital signal (as shown in FIG. 1) is converted into an analog signal, which is recorded on a chip recorder (4), and when played back, the analog signal is converted into a digital signal again. (6) It also has a function to store synthesized sound parameters. (7) is a voice synthesis circuit; when the rotation speed of the chizu recorder (4) during playback is the same as the speed during recording, the output of the A/D converter (6) restores the synthesized sound parameters. As a result, the speech synthesis circuit (7) outputs a speech output of the analog signal. In addition, when the rotational speed of the chizu recorder (4) during playback is twice as high as during recording, every other synthesized sound parameter converted from an analog signal to a digital signal by the A/D converter (5) is are thinned out and synthesized. However, even if the non-stationary part of a consonant is a synthesized sound parameter at the timing of thinning out, the distance from Pn) used for synthesis without thinning out as much as possible (d(n - s >
n) (e.g. Euclidean distance) and calculate the distance 1'
lI d (n-1) The synthesized sound parameter (Pn) is thinned out when n is smaller than a certain value (dth), and is not thinned out when it is not smaller. Therefore, the unsteady part is reproduced as is, and the length of the steady part is shortened.

Note that the system of the present invention requires a speech analysis circuit that performs real-time processing, but it is currently being researched and developed using a signal processing IC, etc., and it is thought that it will be implemented as an IC in the future. Moreover, as a voice synthesis circuit, ICs such as PARCOR type are already commercially available. Therefore, it is believed that the present invention can be constructed inexpensively and with a simple circuit.

〔Effect of the invention〕

The present invention is equipped with a voice analysis circuit that converts the voice input of an analog signal into a digital signal and extracts synthesized sound parameters. A function that calculates the distance between two adjacent synthesized sound parameters, compares this distance with a reference distance, and if it is shorter than the reference distance, does not use the synthesized sound parameter input later in time for voice synthesis. Since it is equipped with a voice synthesis circuit that has a voice synthesis circuit, when audio is played back in fast forward, only the steady part can be shortened while leaving the non-stationary part as much as possible, resulting in a reproduced sound with less deterioration of sound quality and improving the sound quality. have an effect.

[Brief explanation of drawings]

FIG. 1 is a circuit configuration diagram of an embodiment of the present invention, FIG. 2 is an explanatory diagram of the same operation, and FIG. 3 is an explanatory diagram of the operation of the conventional example. +1
1 is an A/D converter, (2) is a voice analysis circuit, (4) is a QI/R, +81 is a buffer, (7
) is a speech synthesis circuit.

Claims (1)

[Claims] (1) Equipped with an A/D converter that converts audio input of analog signals into digital signals, and a speech analysis circuit that extracts synthesized sound parameters from the output of the A/D converter; When making the time axis of audio output shorter than the audio input, the distance between two synthetic sound parameters can be calculated by using two adjacent synthetic sound parameters. Calculate,
A speech synthesis circuit is provided which has a function of comparing this distance with a reference distance and outputting speech without using the synthesized sound parameter input later in time for speech synthesis when the distance is shorter than the reference distance. A recording/playback machine.