JPH04301700A - Digital audio signal reproducing device - Google Patents

Abstract

PURPOSE:To realize fast listening with high magnification such as four times. CONSTITUTION:An address from an address counter 14 is controlled so that sound data of 10 seconds can be read out from sound memory 10 at every 10th second. A pitch that is the repetitive cycle of read out sound data is detected by a pitch detector 22. Since the data is thinned at every pitch by a thinning circuit 30 corresponding to a detected pitch, the data is reduced to 1/2, which reduces the data to 1/4 in total. Thereby, it is possible to listen to recognizable sound intermittently, and to attain the fast listening with high magnification by the recognizable sound.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital audio signal reproducing apparatus suitable for cueing, etc., which reduces the number of data in a digital audio signal and reproduces the reproduced audio signal in a short time.

0002

2. Description of the Related Art Various types of audio recording and reproducing devices have been known. Among these, voice recording and playback devices that use semiconductor memory have the advantage of being able to perform high-speed and random access, and do not require a mechanical drive mechanism for playback. Widely used in recording and reproducing devices.

On the other hand, in audio recording and reproducing apparatuses, recorded data is reproduced at high speed in order to select audio data desired for recording.

In particular, in a playback device that stores digital audio signals in a semiconductor memory, a part of the data is thinned out during digital signal processing to reduce the number of data and playback time is shortened to perform fast listening playback.

[0005]

[Problems to be Solved by the Invention] However, in conventional devices, the limit is to reduce the audio data to about 1/2, and if the data is thinned out further than this, when played back,
There were problems in that it was not possible to understand the content, and furthermore, it was impossible to listen at high speed.

On the other hand, it is preferable for fast listening to be as fast as possible, and it is desired to be able to listen even faster.

The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide an audio reproduction device that allows very high-speed early listening.

[0008]

[Means for Solving the Problems] The present invention provides intermittent sampling means for intermittently sampling an input digital audio signal for a predetermined period of time, and data thinning means for thinning out the sampled audio data to reduce the amount of data. and an output means for continuously outputting the intermittently sampled and thinned out audio data.

[0009]

According to the present invention, the intermittent sampling means samples audio data for each predetermined time period. Data thinning processing is then performed on this intermittently sampled data. Therefore, the amount of data can be reduced by thinning out the sampled data. Therefore, the overall data reduction amount is the sum of intermittent sampling and thinning. Therefore, for example, if you sample for 5 seconds every 10 seconds and thin out 1/2, the amount of data will be 1/4, and 10
This means that seconds worth of audio data can be played back in 2.5 seconds. In other words, 5 seconds worth of audio data is played back in 2.5 seconds, the next 5 seconds worth of data is discarded without being played back, and the next 5 seconds worth of data is played back in 2.5 seconds. will be repeated.

[0010] As described above, if the data thinning rate is about 50%, the data contents after thinning can be heard. Furthermore, if the sampling is performed at a predetermined time interval of about 5 seconds, the playback will be interrupted, but the content can be recognized and it can be sufficiently used for cueing. Therefore, fast listening at a high magnification of about 4 times is possible.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Examples of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of the overall configuration, and audio data is stored in an audio memory 10. This audio memory 10 is a semiconductor memory (RAM), and encodes the input audio signal and stores it as digital data with a reduced amount of data.

This audio memory 10 includes a controller 1
An address counter 14 is connected to the address counter 14 for specifying an access location to the audio memory 10 in response to a signal from the audio memory 10. That is, the controller 12 recognizes modes such as recording, playback, and fast playback, and supplies a predetermined signal to the address counter 14. Therefore, the address counter 14 outputs an address signal indicating the access location in response to a predetermined clock. Then, the audio memory 10 records in the memory at the address specified by the address signal or outputs the contents stored in the memory at the address. For early listening,
In accordance with the address signal output from the address counter 14 every sampling period (for example, every 10 seconds), the contents of the address memory are sequentially output.

On the other hand, a pitch detection buffer 20 is connected to the output path of the audio memory 10.
The data read from 0 is temporarily stored here in predetermined amounts. A pitch detector 22 is connected to the pitch detection buffer 20, and the pitch of the audio data read from the audio memory 10 is detected by the pitch detector 22.

[0014] Here, the pitch refers to the repetition period of the sound, which is determined depending on the type of the sound source. For example, in the case of a human voice, the fundamental frequency is determined by the person's vocal cords, and voices of various frequencies are emitted by various harmonics of this fundamental frequency. Therefore, the sound produced by a normal person repeats its strength and weakness at a period corresponding to the person's fundamental frequency, and this is generally called the pitch period.

Therefore, this pitch period can be detected by determining the autocorrelation between the temporally earlier audio data stored in the pitch detection buffer 20 and the temporally later audio data that is input. Then, by continuously reproducing the audio data for this detected pitch period, it is possible to reproduce a single sound that can be recognized by humans. Note that the pitch period for normal human voice is about 10 to 30 seconds. The pitch period of one person does not change much. Therefore, the pitch may be detected for all data or may be detected intermittently.

Data regarding the pitch period detected by the pitch detector 22 is supplied to a thinning circuit 30. This thinning circuit thins out the audio data supplied from the pitch detection buffer 20 according to the pitch cycle. That is, the detected pitch period continues for 20 msec.
c, after outputting 20 msec of data, the next 20 msec of data is discarded, and this process is repeated to thin out part of the data. Then, if data is discarded for each pitch period according to the pitch detected in this way, the length of the sound will be shortened, but the sound will not be cut in the middle of the pitch, which is the unit of sound, and the shortening will occur. The resulting adverse effect on audio reproduction can be minimized.

Furthermore, the thinning circuit 30 includes a decoding section 4.
0, a FIFO buffer 42 is connected thereto. This decoding section 40 converts the read encoded data into digital audio data similar to A/D conversion of an audio signal. This digital audio data is then output from the FIFO buffer 42 at a predetermined frequency. For this purpose, fs is supplied to the FIFO buffer 42 as a read clock, and this fs is Fs.
It is synchronized with the clock of the D/A converter which converts the digital data from the IFO buffer 42 into an analog signal. Therefore, by A/D converting the signal from the FIFO buffer 42, the audio signal can be reproduced.

FIG. 2 is a flowchart showing the operation of this device during fast listening playback. First, a playback start address is set in the address counter 14 by a command from the controller 12 (S1). As a result, the address counter 14 sequentially increments the addresses starting from the reproduction start address. Therefore, the data stored in the address specified by the address counter 14 is sequentially output from the audio memory 10.

The data read from the audio memory 10 is transferred to the pitch detection buffer 20 (S2), and the pitch is detected by the pitch detector 22 based on this data (S3). The data from the pitch detection buffer 20 is then thinned out based on the pitch in the thinning circuit 30, decoded, and then transferred to the FIFO buffer 42 (S4), from which it is output at a predetermined frequency.

On the other hand, the controller 12 determines whether data equivalent to 10 seconds has been read based on the count value of the address counter 14 (S5), and if 10 seconds have elapsed, the count value placed in the address counter 14 is changed to the value equivalent to 10 seconds. (S6). Therefore, the data for the next 10 seconds will be discarded. Then, if this read data is thinned out to 1/2 in the thinning circuit 30, as shown in FIG.
The sound will be played in seconds.

Furthermore, in the above example, the pitch detector 22
In , pitch was detected by detecting autocorrelation, but pitch can also be detected by other methods. This will be explained based on the flowchart of FIG.

Thinning out in the thinning circuit 30 is turned off (S201). Then, after setting the variable c1 to 0 (S202), in this state, read out one word at a time (S202).
203), it is determined whether the conditions are satisfied (S206). Here, before this condition determination, 1 is added to c1 (S2
04), it is determined whether this c1 has not reached a predetermined constant n (S205). This constant n is a number corresponding to one normal pitch, and is set to correspond to about 10 to 30 msec. Note that this n may be set to a value approximately twice the previous c1.

If the conditions are satisfied and a pitch is detected, thinning is turned on (S207), c1 is set to 0 (S208), and the pitch is detected in the same way (S207).
209-S213). Then, when the pitch is detected, the process returns to S201 and thinning is turned off. by this,
Playback can be alternately switched on and off each time a pitch is detected, and fast listening that shortens the audio can be performed.

Here, the condition determination in S206 and S212 is performed based on whether the following three conditions are satisfied.

(a) a<b (b) a*b<0 (c) |a|+|b|<k/i For example, as shown in FIG. It is something that is repeated. The place where this amplitude becomes small is the break (pitch) of the pitch period. Therefore, when AD converting such an audio signal, the amplitude value at that time is converted into digital data at a predetermined sampling period. Then, this value is stored in the voice memory 10, but if this value is a value for the amplitude centered on 0, n and n output from the voice memory 10 are
The data at the +1 sampling point is an amplitude value having positive and negative signs like a and b shown in the figure. Therefore,
By taking points that satisfy the above-mentioned (a) and (b) for this value, it is possible to extract points where the amplitude value changes from negative to positive. Condition (c) indicates that the amplitude value is less than or equal to a predetermined threshold value, and the pitch can be detected by extracting a predetermined low amplitude point.

[0026] Also, this threshold value is determined as follows for the maximum amplitude k:
It is determined to be a predetermined ratio (for example, 1/16, i=16). Therefore, the determination of this maximum amplitude k will be explained based on FIG. 6. First, at the beginning, the maximum positive value that the audio data can take (for example, if the amplitude is expressed as 8-bit data, the maximum value that can be expressed with 8 bits) is set to k as the initial value. (S10
1). Then, a variable c2 for determining the timing of k update
and set the variable M, in which the maximum value is temporarily input, to 0 (
S102, 103).

Then, data regarding the amplitude is read out and input to variable D (S104). Next, it is determined whether the read data D is larger than the previous maximum value M (S105), and if it is larger, the current data D is input to the variable M to update the maximum value (S106). . Then, 1 is added to the variable c2 (S107), it is determined whether c2>L (S108), and if c2 is less than or equal to L, the process returns to S104 until c2 reaches the maximum value. The process of inputting into M is repeated.

[0028] Here, this L is a predetermined value, and is set to be sufficiently larger than n, which represents approximately one pitch described above. Then, when c2>L, the value of M is substituted for k (S109), and the process returns to S102. Therefore, for every L (eg, approximately pitch), the maximum value at that time is input to k. This value of k will continue for the next six pitches.

Note that, as described above, since k is initially set to the maximum value, the amplitude threshold k/i is a large value. Therefore, it is easy to find points that satisfy the conditions. From the next time onwards, suitable pitch detection can be performed using the actually detected k.

[0030]

As described above, according to the audio reproducing apparatus according to the present invention, sampling at predetermined intervals and thinning are combined to enable fast listening reproduction at high magnification.

[Brief explanation of drawings]

FIG. 1 is a block diagram of the overall configuration of an embodiment.

FIG. 2 is an operation flowchart during fast listening playback.

FIG. 3 is an explanatory diagram of fast listening playback.

FIG. 4 is a flowchart of pitch detection.

FIG. 5 is an explanatory diagram illustrating setting of conditions.

FIG. 6 is a flowchart showing the operation of k calculation.

[Explanation of symbols]

10 Voice memory 12 Controller 14 Address counter 20 Pitch detection buffer 22 Pitch detector 30 Thinning circuit 40 Decoding section 42 FIFO buffer

Claims (1)

[Claims] 1. A digital audio signal reproducing device, comprising intermittent sampling means for intermittently sampling an input digital audio signal for a predetermined period of time, and data for thinning out the sampled audio data to reduce the amount of data. A digital audio signal reproducing device comprising: a thinning device; and an output device that continuously outputs the intermittently sampled and thinned audio data.