JPH0479531A - Method and device for data interpolation - Google Patents

Abstract

PURPOSE:To prevent degradation in the quality of a decoded audio signal by using a weight coefficient extrapolated with the effective value of a first half of a frame just before a frame to be interpolated and an effective value of a latter half of the frame, correcting the amplitude of a segmented waveform and using the result for the interpolation. CONSTITUTION:In the case of receiving metric information 5 from a viterbi decoding section 4, a frame interpolation deciding section 6 decides whether or not an error is present in a viterbi-decoded data and outputs a frame interpolation requiring signal 7. Moreover, an interpolation data generating section 8 generates an interpolation data 9 based on the frame interpolation requiring signal 7 sent from the frame interpolation deciding section 6 and outputs the data. That is, the effective value of the first half of a frame just before a frame to be interpolated and the latter half of the frame is obtained and a weight coefficient depending on the extrapolation of the two effective values is calculated and the weight coefficient is multiplied in the case of repetition of the waveform. Thus, the envelope of the waveform subjected to weighting and interpolation is continuous to the envelope of the waveform of the just preceding frame. Thus, the degradation in the quality of the decoded audio signal is prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a data interpolation method and apparatus, and particularly to an interpolation method for interpolating waveforms in sections that cannot be decoded due to transmission errors in digital transmission of data such as voice data. It is something.

Conventional technology When data signals such as voice are divided into frames of a fixed length and encoded and transmitted digitally, decoding of the data may become difficult due to errors in the transmission path, etc. In such cases,
A frame that has become difficult to decode is replaced with another waveform, that is, interpolation processing is performed. Conventional examples of such data interpolation methods include those shown in FIGS. 4 and 5, for example.

FIG. 4 shows a flowchart of frame-by-frame processing using a conventional data interpolation method, and FIG. 5 shows an example of a waveform interpolated by this method. In Figure 5, the symbol T is 1 frame 1
gnome length and number of samples). Now, the decoded audio sample of the n-th function 1.

Xl, (i), i =0.1.2. ..., T
, -1.

When encoded data for one frame is received in the n-th frame, error detection is performed. If the received data (i) is correct, normal decoding processing is performed and the decoded audio sample Xn (i) is obtained and output.If the received data contains an error that cannot be decoded, th decoded audio sample (i), i = 0.1.2.
..., 'l'', -1 by cutting out TA zambles from the end and repeating them, X, (i) -X,
−, (T, −T, + (i, mod T,) )
i=o, 12. ..., T, ...... (1)
and (7) are output instead of the n-th decoded audio sample. Here, T is calculated using a correlation coefficient to ensure good continuity when repeating the waveform. Figure 5 (a) , (b
) cut out the waveform of partial interval a from interval A, and
15 shows an example in which the interval A' is interpolated by repeating this process.

By using this interpolation method, the audio waveform of frame 1/-, which cannot be decoded, is interpolated with the waveform extracted from the immediately previous frame, thereby maintaining the continuity of the audio.

Invention solves the problem 1. However, in such conventional data interpolation methods, it is difficult to perform good interpolation when the envelope of the audio waveform of the input data signal is stationary, as shown in FIG. 5(a). However, if the envelope of the sound p waveform changes as shown in (1) in Figure 5, the envelope becomes discontinuous as the waveform repeats, degrading the quality of the decoded speech. Cause.

The present invention has been made in view of the above-mentioned problems, and an object thereof is to provide a data interpolation method and apparatus that can perform good interpolation even if the waveform envelope of an input data signal changes.

Means for Solving the Problems In order to achieve the above-mentioned objective 1, the present invention eliminates interpolation and calculates the effective values of the first half waveform and the subsequent waveform of the 71st norm immediately before the 1st norm. , a weighting coefficient determined by extrapolation of these two effective values is calculated 15, and this weighting coefficient is multiplied when the waveform is repeated 1-.

According to the present invention, with the above configuration, the envelope of the weighted and interpolated waveform becomes continuous with the waveform envelope of the immediately previous frame, and an interpolated waveform with good consistency can be obtained.

Embodiment 1 to 3 are diagrams showing an embodiment of the present invention. FIG. 1 is a block diagram of the above embodiment, and FIG.
The figure is a flowchart of frame-by-frame processing using the interpolation method in the embodiment, and FIG. 3 shows an example of a decoded audio waveform after interpolation. Also, in Figure 3, ■, is 1
Freight is the length (number of samples), and T is the length (number of samples) of the cut-out waveform.

In FIG. 1, reference numeral 1- is a data input section into which received data is input, and 2 is a demodulation section which demodulates the received take and outputs demodulated data 3. 4 is a Viterbi decoding unit which decodes the demodulated data 3 using the Viterbi algorithm, stores the branch metric in each state at each time point during decoding, and when decoding the demodulated data at the time point by traceback. The branch metric in the state selected at that time is output as metric information 5. 6
is a frame interpolation determination unit, and 1. When the metric information 5 is received from the Viterbi decoding unit 4, it is determined whether or not there is an error in the Viterbi decoded data based on the metric information 5, and a frame interpolation request signal 7 is output.

8 is an interpolation data creation section, which creates interpolation data 9 based on the frame interpolation request signal 7 sent from the frame interpolation determination section 6. Output. Reference numeral 10 denotes an audio decoding unit, which decodes the received audio coded data 1 outputted from the Viterbi decoding unit 4 and decodes the audio 12 when the frame interpolation request signal 7 is not output in the frame interpolation determining unit 6. It is output from the data output section 3. Further, when the frame interpolation request signal 7 is outputted in the frame interpolation determination section 6, the received audio encoded data 11 of the erroneous frame is not used, and the interpolation data 9 created in the interpolation data creation section 8 is used.
The frame is interpolated using , and audio 12 is output.

The operation of the data interpolation device having such a configuration will be explained below.

In the processing step (hereinafter simply referred to as step) 2 in FIG. The frame interpolation determining unit detects errors in the received data based on the metric information 5 outputted as a result (step 22). In this error detection process, the frame interpolation determining unit 6 checks whether the received data is correct or not (step 23), and if it is correct, the process proceeds to decoding process (step 24) and receives the received audio code output from the Viterbi decoding unit 4. The encoded data 11 is decoded, and the decoded audio sample X, (i), i = 0.12. -・・T, -
1 is obtained and output as voice 12 from the data output section 13. On the other hand, in step 23, if the received data contains an error and cannot be decoded, the immediately preceding (ltl, n-1
decoded audio sample Xl of frame (th), −+ (i), i = 0, L2
, . . . from T, -1 to the length i4 'F as in the conventional example.
Cut out the wave pattern of A (step 25). Next, n-
The effective value e1 of the first half and the effective value e2 of the second half of the first hnome are determined as follows. A weighting coefficient K is determined from e, , e2 and TIl using the following equation (step 26).

Next, the decoded audio sample of the nth frame is interpolated by repeating the extracted waveform while multiplying it by the weighting coefficient () (step 27). For example, according to the embodiment described above, the amplitude of the extracted waveform is corrected using the weighting coefficient extrapolated from the effective value of the first half and the effective value of the latter half of the frame immediately before the frame to be interpolated. Since this is used for interpolation, the waveform envelope of the interpolated frame and the waveform envelope of the immediately previous frame become continuous as shown in Figure 3.
The quality deterioration of the decoded audio that has been subjected to waveform interpolation is reduced.

As described in detail, according to the present invention, 1. Since the amplitude of the interpolated waveform is corrected using a weighting coefficient extrapolated from the effective value of the first half and the effective value of the second half of the immediately preceding frame, the continuity of the envelope of the interpolated audio waveform is improved, making decoding easier. It is possible to suppress audio quality deterioration to a small level and realize faithful audio reproduction.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a data interpolation device according to the present invention, FIG. 2 is a flowchart for implementing the data interpolation method according to the present invention, and FIG. 3 is a waveform diagram showing an example of a decoded audio waveform by the data interpolation process. , FIG. 4 is a flowchart showing an example of a conventional data interpolation method, and FIG. 5 is a waveform diagram showing an example of a decoded audio waveform by the conventional data interpolation method. ]...Data input unit, 2...Demodulation unit, 4...Hitabi decoding unit, 6...Frame interpolation determination unit,...Interpolation data creation unit,...Audio decoding unit,...Data Output section.

Claims (1)

[Claims] 1) When data is divided into frames of a fixed length and encoded and transmitted digitally, when interpolating a frame that is difficult to decode, the effective value of the first half of the previous frame and the effective value of the second half of the previous frame are used. interpolate the amplitude of the waveform used for interpolation by extrapolating the effective value of the first half of the previous frame and the effective value of the second half,
A data interpolation method that makes the waveform envelope of the interpolated frame and the frame immediately before it continuous. 2) A data input section into which received data is input, means for demodulating the received data, means for Viterbi decoding the demodulated data from the demodulating means, and means for audio decoding the output data from the Viterbi decoding means. and means for detecting errors in the decoded data based on the Viterbi-decoded information, and based on the error detection by the error detection means, the effective value of the first half and the second half of the frame immediately before the frame to be interpolated. means for creating interpolated data by correcting the amplitude of the cut out waveform using a weighting coefficient extrapolated from the effective value of;
A data interpolation device consisting of.