JP2004194048A - Transfer method and reproduction method of audio data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a transfer method and reproduction method of audio data, capable of improving the tone quality at the time of data loss. <P>SOLUTION: A lost part 7 where a frame data part 4 is lost in a packet 1 is detected on the basis of the byte size of reception data, the estimation data estimated to be present originally in the detected lost part 7 are estimated on the basis of the frame data part 4 adjacent in front and at the back of the lost part 7 and the lost part 7 is repaired by interpolating the estimated data to the lost part 7. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
[Industrial applications]
The present invention relates to an audio data transfer method and a reproduction method, and more particularly to an audio data transfer method and a reproduction method suitable for sequentially transferring audio data to a receiving side in real time in packet units.
[0002]
[Prior art]
2. Description of the Related Art In recent years, communication technology has been remarkably advanced, and data transferred using a communication line is from a small amount of data such as character data and image data to a large amount of data such as audio data such as music data. Until then, the types are diverse.
[0003]
When transferring data using such a communication line, there is a case where data is divided into packets each having a unit of a fixed size and each packet is sequentially transferred to a receiving side in real time (for example, Patent Documents 1 to 3). 3).
[0004]
Here, as shown in FIG. 9, the unit packet 1 is composed of a packet data section 2 and a packet header section 3 attached to the head of the packet data section 2. Further, the packet data section 2 includes a frame data section 4 consisting of an encoded data section 5 and a frame header section 6 having a predetermined index number attached to the head of the data section 5 in a predetermined order. Is set by arbitrarily setting the number according to.
[0005]
Recently, Bluetooth (Bluetooth) has been adopted as a new communication specification for wirelessly connecting an information portable terminal and a personal computer, and a real-time audio data transfer service using the Bluetooth has been put into practical use. The transfer of one unit of data is performed.
[0006]
However, in real-time audio transfer in Bluetooth, support for Quality of Service (Qos) has not yet been satisfied.
[0007]
For this reason, when the communication line quality is deteriorated, packets that have not been transmitted / received within a unit time have been discarded (packet loss).
[0008]
For example, as shown in FIG. 10, out of the packet data part 2 composed of four frame data parts 4, only the fourth frame data part 4 counted from the head is not received, and the missing part of the frame data part 4 is not received. Consider the case where 7 occurs.
[0009]
In such a case, conventionally, the entire packet 1 is discarded as an incomplete packet, and instead, a dummy packet 8 as shown in FIG. 11 is generated, and the generated dummy packet 8 is stored as a buffer on the receiving side. And temporarily stored it.
[0010]
FIG. 12 is a flowchart showing the procedure of such conventional real-time audio transfer.
[0011]
According to FIG. 12, first, in step 1 (ST1), the receiving side acquires a received packet memory, and then proceeds to step 2 (ST2).
[0012]
In step 2 (ST2), after receiving packet 1 within a unit time from the transmitting side, the process proceeds to step 3 (ST3).
[0013]
In step 3 (ST3), it is determined whether or not the reception of the packet 1 has been completed within the unit time, and if completed, the process proceeds to step 4 (ST4), and if not completed, the process proceeds to step 5 (ST5). .
[0014]
In step 4 (ST4), the received packet 1 is stored in the packet storage and temporarily stored.
[0015]
In step 5 (ST5), after discarding the uncompleted packet 1 in step 3 (ST3), the process proceeds to step 6 (ST6).
[0016]
Then, in step 6 (ST6), the dummy packet 8 is generated, and the dummy packet 8 is stored in the packet storage unit.
[0017]
The packet 1 received by the receiving side and temporarily stored is decoded by a subsequent decoder and reproduced as audio.
[0018]
[Patent Document 1]
JP-A-11-205291 (FIG. 1)
[Patent Document 2]
JP-A-11-331261 (FIG. 1)
[Patent Document 3]
Japanese Patent Publication No. 2000-507777 (FIG. 1)
[0019]
[Problems to be solved by the invention]
By the way, the dummy packet 8 that replaces the discarded incomplete packet usually has zero-value data interpolated in each frame data section 4, and as shown in FIG. In some cases, the data portions (FD1 to FD4) obtained as zero value data.
[0020]
Therefore, in the received audio data, the packet loss, that is, the portion where the packet 1 is discarded, is in a state where there is substantially no data portion to be reproduced even if this portion is supplemented by the dummy packet 8. .
[0021]
As a result, in the related art, when audio data is reproduced, a portion where a packet loss occurs is clearly interrupted so that it can be easily detected by audibility, and the reproduced sound quality is significantly deteriorated.
[0022]
The present invention has been made in view of the above problems, and has as its object to provide an audio data transfer method and a reproduction method that can improve sound quality at the time of data loss.
[0023]
[Means for Solving the Problems]
To achieve the above object, the feature of the audio data transfer method according to the present invention is to detect a missing part where reception of a frame data part in a packet is missing based on the byte size of the received data, and to detect the missing part. The estimated data that originally existed is estimated based on frame data portions adjacent before and after the missing portion, and the estimated portion is interpolated into the missing portion to repair the missing portion. On the point.
[0024]
By adopting such a method, the missing portion can be repaired by interpolating the estimated data into the missing portion, and as a result, audio data with good sound quality without sound interruption can be reproduced. In addition, since repair can be performed in units of frames before decoding instead of in units of decoded audio samples, the amount of processing can be reduced, and missing data can be eliminated by filtering during decoding performed at a later stage. It is possible to reproduce audio data having a smoother sound quality than the estimated data inserted in the section.
[0025]
Further, the feature of the audio data transfer method according to the present invention is that, when a plurality of missing portions consecutive in front and back occur, a frame data portion adjacent to the front of the leading missing portion and a trailing portion of the trailing missing portion. The point is that the estimation data is estimated based on adjacent frame data portions.
[0026]
By adopting such a method, even when a plurality of missing portions are continuous, it is possible to estimate the estimation data and repair the missing portions.
[0027]
Further, a feature of the audio data transfer method according to the present invention resides in that the transfer of the audio data is performed using Bluetooth.
[0028]
By adopting such a method, it is possible to improve the sound quality at the time of packet loss even when audio data is distributed using Bluetooth having a dissatisfied QoS.
[0029]
Still further, a feature of the audio data transfer method according to the present invention is that a data pattern corresponding to each of frame data portions adjacent before and after the missing portion and estimated data corresponding to each set of data patterns are stored. The missing part is repaired using a look-up table.
[0030]
Then, by adopting such a method, estimated data can be easily obtained, and repair of a missing portion can be quickly performed.
[0031]
Also, the feature of the audio data reproducing method according to the present invention is that a missing part where reception of the frame data part in the packet is lost is detected based on the byte size of the received data, and the detected missing part originally exists. The estimated data that would have been lost is estimated based on the frame data portions adjacent before and after the missing portion, and the estimated data is interpolated into the missing portion to repair the missing portion. The point is to play back the packet that was created.
[0032]
Then, by adopting such a method, the missing data can be repaired by interpolating the estimated data into the missing data, and as a result, a good reproduced sound quality without sound interruption can be obtained. In addition, since the repair can be performed in units of frames before decoding instead of in units of decoded audio samples, the processing amount can be reduced, and the filtering during the decoding process performed in the subsequent stage can reduce Audio data having a sound quality smoother than that of the inserted estimation data can be reproduced.
[0033]
Furthermore, the feature of the audio data reproducing method according to the present invention is that, when a plurality of missing portions continuous in front and back occur, a frame data portion adjacent to the front of the leading missing portion and a trailing portion of the trailing missing portion. The point is that the estimated data is estimated based on the adjacent frame data portion, and the estimated data is inserted into the missing portion to reproduce the repaired packet.
[0034]
By adopting such a method, even when a plurality of missing portions are continuous, it is possible to estimate the estimation data and repair the missing portions.
[0035]
Furthermore, a feature of the audio data reproducing method according to the present invention resides in that the transfer of the audio data is performed using Bluetooth.
[0036]
By adopting such a method, it is possible to improve sound quality at the time of packet loss and obtain good reproduced sound quality even when audio data is distributed using Bluetooth having unsatisfactory QoS. .
[0037]
Also, a feature of the audio data reproducing method according to the present invention is that a data pattern corresponding to each of frame data portions adjacent before and after the missing portion and estimated data corresponding to each set of data patterns are stored. The point is that the missing portion is repaired using a look-up table, and the repaired packet is reproduced.
[0038]
Then, by employing such a method, estimated data can be easily obtained, and audio data with the missing portion repaired can be simply and quickly reproduced.
[0039]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an embodiment of a method of transferring and reproducing audio data according to the present invention will be described with reference to FIGS.
[0040]
It should be noted that the same or similar parts as those of the related art will be described using the same reference numerals.
[0041]
FIG. 1 shows an example of a transfer system 10 for implementing a method of transferring and reproducing audio data according to the present embodiment.
[0042]
The transfer system 10 includes two communication devices, such as a transmitter 11 and a receiver 12, each of which is composed of a personal computer or a portable terminal device for transmitting and receiving audio data in packet units.
[0043]
The receiver 12 has a packet storage unit 13 such as a buffer for temporarily storing the received packet 1, and the packet storage unit 13 includes, as shown in FIG. A series of packets 1 before and after transmission are sequentially received and accumulated.
[0044]
If the packet 1 stored in the packet storage unit 13 is received in a complete state without any loss of the frame data unit 4, the packet 1 is delivered to the subsequent decoder unit 14 as it is.
[0045]
On the other hand, as shown in FIG. 2, when the reception of the frame data portion 4 in the packet 1 is lost, a lost portion 7 occurs. Conventionally, the entire packet in which the lost portion 7 occurred is discarded as an incomplete packet. However, in the present embodiment, a complete packet 1 is created by repairing the missing part 7.
[0046]
That is, the packet storage unit 13 is connected to a control unit 16 that controls a repair process for the missing unit 7. The control unit 16 includes a byte of data of the packet 1 received by the packet storage unit 13. A byte number measuring unit 17 for counting the number is connected.
[0047]
The control section 16 detects the position and the number of the missing portions 7 of the frame data section 4 based on the measurement result of the byte number measuring section 17.
[0048]
Further, after detecting the missing portion 7, the control unit 16 calculates the estimated data as the frame data portion 4 which would have originally existed in the missing portion 7, in a frame adjacent to the frame before and after the missing portion 7. The estimation is performed based on the data section 4.
[0049]
As means for determining the estimation data, an estimation data storage unit 18 such as a RAM is connected to the control unit 16. For example, as shown in FIG. A lookup table 19 in which frame data portions 4 adjacent before and after the missing portion 7 and estimated data corresponding to each frame data portion 4 are stored.
[0050]
Incidentally, the the look-up table 19 configured will be described in more detail in, as shown in FIG. 3, wherein the look-up table 19, the data pattern corresponding to the frame data unit 4 adjacent to the front of the missing portion 7 pt _- a (Pt1,..., PtA,..., PtN) and a data pattern pt _- b (Pt1,..., PtB,...) corresponding to the frame data part 4 adjacent after the missing part. Are stored respectively. Further, in the lookup table 19, a plurality of estimated data (ed11,...) Corresponding to a pair of data patterns (pt _- a, pt _- b) adjacent before and after the missing portion 4 are stored. Have been.
[0051]
When repairing the missing part 7, the control unit 16 searches the lookup table 19 for estimated data corresponding to the frame data part 4 adjacent before and after the missing part 7. I have. That is, the control unit 16 firstly selects one of the plurality of data patterns pt _- a (for example, PtA) corresponding to the frame data part 4 adjacent to the frame part 4 immediately before the missing part 7; Among the plurality of data patterns pt _- b, one of the data patterns (for example, PtB) corresponding to the frame data portion 4 that is actually adjacent after the missing portion 7 is estimated and selected. . Further, the control unit 16 is configured to estimate the estimated data (edAB in the above example) corresponding to each set of the estimated data patterns (for example, PtA and PtB).
[0052]
Then, the missing data 7 is repaired by interpolating the retrieved estimated data into the missing data 7 of the incomplete packet stored in the packet storage unit 13.
[0053]
The look-up table 19 stored in the estimation data storage unit 18 need not be limited to one, and a plurality of look-up tables 19 may be stored as needed.
[0054]
For example, when a plurality of missing portions 7 connected in front and behind occur, the following three lookup tables 19 are stored in the estimated data storage unit 18 in order to repair the plurality of missing portions 7. Is also good.
[0055]
That is, first, as shown in FIG. 4, the first look-up table 19 includes a frame data part 4 adjacent to the front side (FDf in FIG. 4) and a frame data part 4 adjacent to the rear side (FDf in FIG. 4). FDr) is a look-up table 19 (LUT1) used to search for estimation data corresponding to the target missing portion 7 located at the center when the number of missing portions 7 present between the FDr) is odd.
[0056]
Next, as shown in FIG. 5, the second lookup table 19 includes a frame data part 4 adjacent to the front side (FDf in FIG. 5) and a frame data part 4 adjacent to the rear side (FDr in FIG. 5). ), The estimated data corresponding to the noticeable missing portion 7 located on the left side immediately adjacent to the center position (in this case, the boundary of the missing portion 7) is determined. It is a lookup table 19 (LUT2) used for searching.
[0057]
Finally, as shown in FIG. 6, the third look-up table 19 includes a frame data section 4 adjacent to the front side (FDf in FIG. 6) and a frame data section 4 adjacent to the rear side (FDr in FIG. 6). )), The lookup table 19 used for searching for the estimated data corresponding to the noticeable missing portion 7 located on the right side nearest to the center position (boundary) when the number of the missing portions 7 existing between them is an even number. (LUT3).
[0058]
Estimated data obtained using any of these three look-up tables 19 (LUTs 1 to 3) can be used as a frame data section 4 for repairing the remaining missing section 7. For example, when five missing parts 7 exist between the frame data part 4 adjacent on the front side and the frame data part 4 adjacent on the rear side, first, the center (third from the top) is determined by the LUT 1. A complete frame data part 4 is obtained by interpolating the estimated data into the missing part 7 of.
[0059]
Then, between the frame data portion 4 obtained by repairing the missing portion 7 and the frame data portion 4 adjacent to the front side, two missing portions 7 still remain.
[0060]
However, the two missing portions 7 are respectively provided with the frame data portion 4 adjacent to the front side and the frame data portion 4 obtained by repairing the missing portion 7 before and after the two missing portions 7. The correction can be made by using the LUT2 and LUT3 by treating them as the adjacent frame data section 4.
[0061]
Similarly, between the frame data portion 4 obtained by repairing the missing portion 7 and the frame data portion 4 adjacent on the rear side, two missing portions 7 still remain, but these two missing portions 7 remain. The missing portion 7 can be repaired by using the LUT2 and the LUT3.
[0062]
As described above, even when a plurality of missing portions 7 are generated, the starting point is the frame data portion 4 adjacent to the front of the leading missing portion 7 and the adjacent missing portion 7 to the rear of the trailing missing portion 7. The repair can be performed successively by appropriately using a plurality of lookup tables 19.
[0063]
When audio data is transferred in real time using the transfer system 10 configured as described above, first, the audio data is transmitted from the transmitter 11 to the receiver 12 in accordance with a predetermined communication format such as Bluetooth. The audio data is transmitted in units of one packet, and the receiver 12 receives each packet 1 and sequentially stores it in the packet storage unit 13.
[0064]
If an incomplete packet occurs as shown in FIG. 2, a missing portion 7 into which data is to be inserted is detected as shown in step 1 (ST1) in FIG.
[0065]
The missing portion 7 is detected by measuring the number of bytes of the data received in the packet storage unit 13 by the byte number measurement unit 17 and based on the measurement result. That is, when the number of bytes of the received packet 1 is less than the number of bytes that the unit packet 1 normally has, the portion where the number of bytes of the received packet 1 is not accumulated is recognized as the missing portion 7.
[0066]
Next, as shown in step 2 (ST2), the frame data part 4 adjacent to the missing part 7 before and after the missing part 7 is detected.
[0067]
In FIG. 2, the last frame data portion 4 in the preceding packet 1 of the two consecutive packets 1 is the missing portion 7. In this case, the preceding packet to which the missing portion 7 belongs is used. 1, the frame data part 4 adjacent to the front of the missing part 7 and the leading frame data part 4 of the packet 1 on the rear side become the frame data parts 4 adjacent before and after the missing part 7.
[0068]
Next, in step 3 (ST3), the lookup table 19 is used to select estimated data to be interpolated in the missing portion 7 based on the frame data portions 4 adjacent before and after the missing portion 7.
[0069]
When the processing in step 3 (ST3) is further subdivided, first, as shown in FIG. 8, in step 3-1 (ST3-1), the data pattern of the frame data section 4 adjacent to the missing part 7 before and after the missing part 7 is obtained. Estimate each value. Next, in step 3-2 (ST3-2), the lookup data is searched from the lookup table 19 using the two estimated data pattern values.
[0070]
Then, in step 4 (ST4), the repair of the missing portion 7 is completed by interpolating the estimated data into the missing portion 7.
[0071]
The complete packet in which the missing portion 7 has been repaired as described above is transferred to the subsequent decoder unit 14, and then decoded by the decoder unit 14 and reproduced as audio.
[0072]
Therefore, in the present embodiment, even when the reception of the frame data portion 4 in the packet 1 is lost, the incomplete packet in which the reception is lost can be discarded as in the related art. Instead, it is possible to effectively utilize the frame data portion 4 properly received among the incomplete packets.
[0073]
The repair of the incomplete packet can be performed simply and appropriately by referring to the look-up table 19, and the audio data with good sound quality without sound interruption can be reproduced.
[0074]
Further, since the data loss can be corrected not in the decoded audio sample unit but in the frame unit before decoding, the processing amount can be reduced, and the missing part 7 can be reduced by the filtering during the decoding process performed in the subsequent stage. It is possible to reproduce audio data having a smoother sound quality than the estimated data interpolated into the audio data.
[0075]
In particular, when the transfer of the audio data is performed by using a communication mode in which QoS such as Bluetooth is unsatisfactory, a more excellent effect can be exhibited in improving the sound quality at the time of data loss.
[0076]
The present invention is not limited to the above-described embodiment, but can be variously modified as needed.
[0077]
【The invention's effect】
As described above, according to the transfer method and the reproducing method of the audio data according to the present invention, it is possible to improve sound quality at the time of data loss, and to obtain good reproduced sound quality.
[0078]
Further, according to the method of transferring and reproducing audio data according to the present invention, it is possible to flexibly cope with a missing portion that may occur in various modes, and it is possible to obtain better reproduction sound quality.
[0079]
Further, according to the method of transferring and reproducing audio data according to the present invention, real-time transfer of audio data can be appropriately performed by Bluetooth.
[0080]
Furthermore, according to the audio data transfer method and the reproduction method according to the present invention, the sound quality can be improved more easily and quickly when data is lost.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an example of a communication system for implementing an audio data transfer method and a reproduction method according to the present invention. FIG. 2 is a block diagram showing an audio data transfer method and a reproduction method according to an embodiment of the present invention. FIG. 3 is a diagram showing a series of packets connected before and after being received by a receiver. FIG. 3 is a schematic diagram showing an example of a look-up table in an embodiment of an audio data transfer method and a reproduction method according to the present invention. FIG. 5 is an explanatory diagram showing a repair method when a plurality of missing portions occur in the embodiment of the audio data transfer method and the playback method according to the present invention. FIG. 5 is a diagram showing the audio data transfer method and the play method according to the present invention. FIG. 6 is an explanatory view different from FIG. 4 showing a repair method when a plurality of missing portions occur in the embodiment. FIG. 6 is an audio data according to the present invention. FIG. 7 is an explanatory view showing a repair method when a plurality of missing portions are generated in the embodiment of the transfer method and the play method different from FIGS. 4 and 5. FIG. 7 is a transfer method and a play method of audio data according to the present invention. FIG. 8 is a flow chart showing a part of FIG. 7; FIG. 9 is a view showing a unit packet forming audio data; FIG. 10 is a view showing a packet data section forming a unit packet. FIG. 11 is an explanatory diagram showing a configuration of a packet data portion of a dummy packet. FIG. 12 is a flowchart showing an example of a conventional audio data transfer method.
1 packet 2 packet data part 3 packet header part 4 frame data part 5 encoded data part 6 frame header part 7 missing part 19 look-up table

Claims (8)

A packet data portion in which an arbitrary number of frame data portions each including an encoded data portion and a frame header portion attached to the head of the data portion are collected in a predetermined order; and a packet attached to the head of the packet data portion. A method of transferring audio data in which a packet of audio data including a header portion is transferred in real time to a receiving side,
The missing part where the reception of the frame data part in the packet is missing is detected based on the byte size of the received data, and the estimated data that would originally exist in the detected missing part is detected before and after the missing part. A method for transferring audio data, comprising estimating based on an adjacent frame data part, and interpolating the estimated data into the missing part to repair the missing part. When a plurality of missing portions connected in front and behind occur, the estimation data is estimated based on the frame data portion adjacent to the front of the leading missing portion and the frame data portion adjacent to the trailing portion of the trailing missing portion. The method of transferring audio data according to claim 1, wherein: The audio data transfer method according to claim 1, wherein the transfer of the audio data is performed using Bluetooth. The missing portion is repaired using a look-up table in which data patterns respectively corresponding to frame data portions adjacent before and after the missing portion and estimated data corresponding to each data pattern set are stored. 4. The audio data transfer method according to claim 1, wherein: A packet data portion in which an arbitrary number of frame data portions each including an encoded data portion and a frame header portion attached to the head of the data portion are collected in a predetermined order; and a packet attached to the head of the packet data portion. A method of reproducing audio data for transmitting a packet of audio data including a header portion to a receiving side in real time and reproducing the transferred packet,
The missing part where the reception of the frame data part in the packet is missing is detected based on the byte size of the received data, and the estimated data that would originally exist in the detected missing part is detected before and after the missing part. A method for reproducing audio data, comprising estimating based on an adjacent frame data portion, repairing the missing portion by interpolating the estimated data into the missing portion, and reproducing the repaired packet. . When a plurality of missing portions consecutive in front and back occur, the estimation data is estimated based on a frame data portion adjacent to the front of the leading missing portion and a frame data portion adjacent to the trailing side of the trailing missing portion. 6. The audio data reproducing method according to claim 5, wherein the repaired packet is reproduced by interpolating the estimated data into the missing portion. The audio data reproducing method according to claim 5, wherein the transfer of the audio data is performed using Bluetooth. The missing portion is repaired using a look-up table in which data patterns respectively corresponding to frame data portions adjacent before and after the missing portion and estimated data corresponding to each data pattern group are stored. 8. The audio data reproducing method according to claim 5, wherein the reproduced packet is reproduced.

Images