JP2008225234A - Audio signal processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To make it possible to specify the start position of an information addition processing frame of audio data relatively easily even when the audio data is input in stream data format by adding information so that the start position of the information addition processing frame of the audio data can be specified relatively easily. <P>SOLUTION: The audio signal processor includes a synchronism information adding means 304 of adding synchronism information showing time positions of frames to an audio data input by the frames as sections of fixed length, and an information adding means 305 of embedding relative information associated with the audio data input in the frames of the audio data with the added synchronism information. Further, the audio signal processor includes a synchronism information detecting means 702 of inputting the audio data having the information embedded in the stream data format and detecting the synchronism information to detect the frame start positions, and a relative information extracting means 704 of processing the audio data having the information embedded on the basis of the synchronism information detection result by the frames as specified and extracting the relative information embedded in the frames. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an audio signal processing apparatus, and more particularly to an apparatus for embedding or extracting synchronization information indicating a time position of a processing frame with respect to audio data and related information related to audio data. Used in the distribution system.

  Digitized music content, eg CD quality digital audio data, consists of 44100 16-bit samples per second for each channel. When digital watermark information such as copyright information or other information is added (embedded) to such uncompressed digital audio data, as a basic method of digital signal processing, samples are sampled in a certain length section. A method of extracting and performing individual processing on the extracted sample is often used. Here, each extraction section having a certain length is called a frame. As a typical technique using frame processing, high-speed Fourier transform can be mentioned.

  Conventionally, an information adding device for adding information to digital audio data of CD quality delimits audio data every 2048 samples, for example, extracts samples in the divided frames, and performs a predetermined information adding process. And output as audio data with added information. Then, the information extraction device to which the information-added audio data as described above is input extracts samples in the frame interval for every 2048 samples that is the same as the operation in the information addition device, and performs predetermined signal processing. Extract information embedded in audio data.

  By the way, in the conventional information extraction apparatus, when the audio data is read from a predetermined recording medium or the like, the start position of the frame of the audio data determined by the operation of the information adding apparatus can be clearly reproduced. However, when audio data is input to the information extraction device in the stream data format, it is generally difficult to specify the start position of the information addition processing frame regardless of whether the input data is analog / digital.

In addition, in order to specify the start position of the information addition processing frame, a method of extracting a specific code while shifting the start position of the frame for each sample or performing an extraction process based on the specification of the frequency of the signal in the frame Has been proposed (see Patent Document 1), however, there is a problem in that the amount of calculation accompanying extraction becomes enormous.
JP 2000-172282 A

  The present invention has been made to solve the above-mentioned conventional problems, and an audio signal that can add information so that the start position of an information addition processing frame of audio data can be specified relatively easily. An object is to provide a processing apparatus.

  Another object of the present invention is to provide an audio signal processing apparatus capable of relatively easily specifying the start position of an information addition processing frame of audio data even when the audio data is input in a stream data format. Is to provide.

  According to a first aspect of the audio signal processing apparatus of the present invention, there is provided synchronization information adding means for adding synchronization information indicating a time position of a frame for each frame which is a fixed length section with respect to audio data input; And related information adding means for embedding related information related to the audio data in the frame of the audio data to which the synchronization information is added by the adding means.

  According to a second aspect of the audio signal processing device of the present invention, predetermined processing is performed for each frame that is a fixed-length section for audio data input, and the time position of the frame is set for the audio data. Synchronization information detecting means for detecting the synchronization information by inputting the embedded audio data in the stream data format in which related information related to the audio data is embedded in the frame. A frame extracting unit having a function of detecting a frame start position of the audio data embedded with the information based on the synchronization information detected by the synchronization information detecting unit, and based on the frame information detected by the frame extracting unit. A predetermined process is performed for each frame of the audio data with the information embedded and embedded in the frame. Characterized by comprising the additional information extracting means for extracting relevant information are, the.

  As the synchronization information, a predetermined low frequency band which is a triangular wave signal having a waveform in which two adjacent frames of the audio data are synchronized with the frame as one cycle and whose amplitude changes in a triangular shape and has little influence on the sense of hearing. It is desirable to use this signal.

  According to the audio signal processing apparatus of the present invention, it is possible to add information so that the start position of the information addition processing frame of audio data can be specified relatively easily. Further, according to the audio signal processing apparatus of the present invention, even when audio data is input in the stream data format, the start position of the information addition processing frame of the audio data can be specified relatively easily.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In this description, common parts are denoted by common reference numerals throughout the drawings.

<First Embodiment>
FIG. 1 schematically shows the configuration of an audio content distribution system according to the first embodiment of the audio signal processing apparatus of the present invention. On the transmission side 10 of this audio content distribution system, audio data of the audio content distribution system is input, an information adding device 30 for adding predetermined information, and a transmission for sending the output of this information adding device 30 to some transmission line 40 Circuit 20. On the receiving side 50 of the audio content distribution system, a receiving circuit 60 that receives an input from the transmission path 40, an information extracting device 70 that extracts predetermined information from the output of the receiving circuit 60, and extracts audio data; It comprises. The information adding device 30 and the information extracting device 70 can be realized by software or hardware, but in this example, they are realized by using a CPU and a program.

  FIG. 2 is a block diagram showing an example of a processing flow (step) focusing on the processing function in the information adding device 30 in FIG. The information adding device 30 includes a frame position specifying unit 302, a frame extracting unit 303, a synchronization information adding unit 304, and a related information adding unit 305.

  The frame position specifying unit 302 receives audio data (audio data) of the audio content distribution system, for example, CD quality audio data, and divides the audio data into frames (for example, 2048 samples) that are constant length sections. It has a function of specifying a frame position (for example, adding frame synchronization information). The frame extracting unit 303 extracts a sample in the frame for each frame divided by the frame position specifying unit 302.

  The synchronization information adding unit 304 has a function of adding synchronization information indicating a time position of a frame to a sample in the frame for each frame of audio data input. This example has a synchronization signal generation function that generates a synchronization signal synchronized with a frame of audio data input, and an addition function that adds (superimposes) the synchronization signal to the audio data input in an analog manner.

  The related information adding unit 305 embeds (adds) the related information related to the audio data in the frame of the audio data to which the synchronization information is added by the synchronization information adding unit 304 according to a predetermined rule, thereby adding information processing. It has a function of outputting as finished audio data. Here, the related information related to the audio data is, for example, text information such as lyrics, copyright information, performance information such as a score or rhythm, performance control information of a performance robot or toy, and the like, for example, of the MIDI standard Added in data format.

  Next, an example of processing in the synchronization information adding means 304 in FIG. 2 will be described with reference to FIG. 3, FIG. 4, and FIG. FIGS. 3A and 3B show examples of signal waveforms of audio data inputted to the frame position specifying means 302 in FIG. Here, the CD quality audio data input in which the sampling value of the amplitude of the audio signal is digitally expressed is taken as an example, and the waveform is expressed by a waveform in which the sampling values are continuously connected. In this example, a left channel L signal and a right channel R signal of stereo data are shown. 4A and 4B show an example of a signal waveform of a frame synchronization signal added as synchronization information to the original audio data. FIGS. 5A and 5B show examples of signal waveforms of audio data with information added by adding a frame synchronization signal to the original audio data.

  The conditions required for the frame synchronization signal are as follows.

  (1) The frame synchronization signal is an AC wave that is synchronized with a frame with two adjacent frames of audio data as one period. In other words, the amplitude of the frame synchronization signal corresponding to both end positions of the frame is the reference level (0), the amplitude is the maximum value or the minimum value at the center of the frame, and the amplitude direction is symmetrical between the temporally adjacent frames. It is an alternating wave that reverses alternately so that This AC wave may be a triangle, a rectangular wave, a sine wave, a triangular pulsating wave having a waveform whose amplitude changes in a triangular shape, etc., but the boundary of the frame is clearly defined when detecting a synchronization signal as described later. It is desirable to use a triangular wave signal for detection.

  (2) As the frame synchronization signal, it is desirable to use a signal in a predetermined low frequency band that has little influence on the audibility when audio data is reproduced, for example, a signal of several tens Hz or less in a band close to a direct current component.

  (3) When a frame synchronization signal is added to stereo data, the amplitude direction of the frame synchronization signal is symmetrical between the left channel frame and the right channel frame of the stereo data at the same time. Is added as follows. Note that this condition (3) is not necessary when a frame synchronization signal is added to monaural data.

  According to the information addition device 30 described above, information can be added so that the start position of the information addition processing frame of the audio data can be specified relatively easily. The information adding device 30 described above can be provided not only on the transmission side of the audio content distribution system as described above but also in any audio signal processing system.

  FIG. 6 is a block diagram showing an example of a processing flow (step) focusing on the processing function in the information extracting device 70 in FIG. This information extraction apparatus includes synchronization information detection means 702, frame extraction means 703, and related information extraction means 704.

  That is, as described above, the synchronization information detection unit 702 performs predetermined processing for each frame that is a fixed-length section on the original audio data, and indicates the time position of the frame for the audio data. The synchronization information is added, and the embedded audio data in which the related information related to the audio data is embedded in the frame is input in the stream data format, and the synchronization information is detected from the audio data input.

  The frame extraction means 703 has a function of detecting the synchronization position (frame start position) every 2048 samples, which is the same as when the information adding device 30 is operated, based on the synchronization information detected by the synchronization information detection means 702.

  The related information extracting unit 704 extracts a sample in each frame section from the audio data in which information is embedded based on the frame information detected by the frame extracting unit 703, performs a predetermined process, and embeds it in the frame It has a function to extract related information.

Next, an example of processing in the synchronization information detecting means 702 in FIG. 6 will be described with reference to FIGS. First, the synchronization information detecting means 702 calculates the transition of the sum of the frame lengths for each channel of audio data input with information embedded. The result y (n) of this calculation is expressed by the following (Equation 1), where x (n) is the input signal and N is the detected frame length.

  Here, FIG. 7A and FIG. 7B show an example of transition data of the sum value of the frame length for the left channel L signal and the right channel R signal of the stereo data.

  Next, the difference (absolute value) between the calculation result yL (n) of the transition of the total value of the left channel L and the calculation result yR (n) of the transition of the total value of the right channel R is calculated. The difference calculation result z (n) is expressed by the following (Equation 2).

z (n) = | yL (n) −yR (n) | (Formula 2)
Here, an example of the inter-channel difference data of the frame total value transition is shown in FIG.

  Next, by detecting the position where the difference calculation result z (n) takes the maximum value, the boundary of the frame to be detected (the start position of the frame) is detected. Here, when the triangular wave signal as described above is used as the synchronization signal, the boundary of the frame can be detected very clearly.

  Next, an example of a processing procedure performed by the synchronization information detecting unit 702 in FIG. 6 will be described in detail. If the detected frame length is N, (A) 0th sample (extraction start point) to (N-1) th sample, (B) Nth sample to (2N-1) th sample And (C) the processing for the 2Nth and subsequent samples.

  FIG. 9 is a flowchart showing an example of the process (A) for the 0th sample (extraction start point) to the (N-1) th sample in the synchronization information detecting means 702 in FIG. In step S1000, initialization is performed with the number of times n = 0, the calculation result YL = 0 of the transition of the total length of the frame length of the left channel L, and the calculation result YR = 0 of the transition of the total value of the right channel R.

  In this state, since the number of samples that can be obtained after the extraction start point is less than N samples, the calculation of (Equation 1) cannot be satisfied. Therefore, in order to evaluate the result of (Equation 1) after N samples, in step S1001, the obtained input samples XL (n) and XR (n) are used as delay buffers x'L (n) and x'R (n ). Next, in step S1002, operations YL = YL + XL (n) and YR = YR + XR (n) for adding the latest input sample XL (n) to the accumulated value are performed. In step S1003, n is incremented. In step S1004, a determination is made. If n <N, the process returns to step S1001.

  FIG. 10 is a flowchart showing an example of the process (B) for the Nth sample to (2N-1) th sample in the synchronization information detecting means 702 in FIG. In step S1101, the obtained input samples XL (n) and XR (n) are stored in the delay buffers x'L (n) and x'R (n). In step S1102, an operation YL = YL-XL (n-N) + XL (n) is performed in which the value XL (n-N) before N samples is removed from the accumulated value YL and the latest input sample XL (n) is added. Further, an operation YR = YR−XR (n−N) + XR (n) is performed by adding the latest input sample XR (n) by subtracting the value XR (n−N) N samples before the accumulated value YR.

  As a result, the value obtained by performing the calculation in step S1103 is equal to the value of (Equation 1) at the time of n samples calculated corresponding to the left channel L and the right channel R. Thus, the calculation of (Equation 1) only needs to be added and subtracted one by one, and the processing load can be kept low.

  Next, the calculation of (Equation 2) is performed in step S1104, and in step S1105, n in which z (n) has the maximum value in the Nth sample to the (2N-1) th sample in this state is identified. Save to variable nf. In step S1106, n is incremented. In step S1107, a determination is made. If n <2N, the process returns to step S1101.

  The sample position stored in nf at the time of completion of this state indicates the start position of the detection frame. Specifically, nf + N, nf + 2N, nf + 3N, ... are used as frame start position samples. can do.

  FIG. 11 is a flowchart showing an example of the process (C) for the 2Nth sample and thereafter in the synchronization information detecting means 702 in FIG. In this state, correction processing is performed on the frame start position obtained in the previous state. The processing from step S1201 to step S1204 is equal to the processing from step S1101 to step S1104 described above. In step S 1205, as in step S 1104, n 2 in which z (n) has the maximum value after the 2N-th sample is identified and stored in the variable nf ′. As a result of the identification, in step S1206, it is detected whether n′f is deviated from the previously obtained nf. If there is a deviation, nf is updated in step S1207. In step S1208, n is incremented, and the process returns to step S1201.

  According to the information extracting device 70 described above, even when audio data is input in the stream data format, the start position of the information addition processing frame of the audio data can be specified relatively easily.

  The present invention is applicable to the field of using an audio portion of a video signal such as a DVD or a movie file and embedding an auxiliary signal such as a control signal in the audio portion. In addition, audio data recorded on CDs or uncompressed audio such as wave formats handled by personal computers, audio compression / decompression technologies such as MP-3 and AAC, such as DVD and music distribution, portable digital music players, and Chaku-Uta It is applicable to the field of transmitting and distributing audio signals such as package and file format using file, or file distribution and streaming. For example, sales of music contents with lyrics (explanation and artist message) display such as Chaku-Uta, karaoke service using audio compression / decompression technology such as MP-3 / AAC, music display service for displaying lyrics, commercialization of musical instrument playing robot toys, It can be applied to fields such as sales of music content with performance information and sales of music content with beat timing.

1 is a block diagram schematically showing the configuration of an audio content distribution system according to a first embodiment of an audio signal processing device of the present invention. The block diagram shown according to an example of a processing flow paying attention to the processing function in the information addition apparatus in FIG. FIG. 3 is a waveform diagram showing an example of audio data input processed by the synchronization information adding means in FIG. 2. FIG. 3 is a waveform diagram showing an example of a frame synchronization signal added as synchronization information to the original audio data by the synchronization information adding means in FIG. 2. FIG. 3 is a waveform diagram showing an example of audio data that has been subjected to information addition processing and has been processed by the synchronization information addition means in FIG. 2; The block diagram shown according to an example of a processing flow paying attention to the processing function in the information extraction device in FIG. FIG. 7 is a characteristic diagram showing an example of a result of calculating the transition of the sum of frame lengths for each channel when the audio data input in which information is embedded by the synchronization information detection unit in FIG. 6 is stereo data. FIG. 6 is a characteristic diagram showing an example of the result of calculating the absolute value of the difference between the transition data of the sum of the frame lengths for each channel when the audio data input in which information is embedded by the synchronization information detecting means in FIG. 6 is stereo data. . The flowchart which shows an example of the process with respect to the 0th sample (extraction start point)-(N-1) th sample in the synchronous information detection apparatus in FIG. The flowchart which shows an example of the process with respect to the Nth sample-(2N-1) th sample in the synchronous information detection apparatus in FIG. The flowchart which shows an example of the process with respect to the 2N-th sample and after in the synchronous information detection apparatus in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Transmission side of audio | voice content delivery system, 20 ... Transmission circuit, 40 ... Transmission path, 30 ... Information addition apparatus, 50 ... Reception side of audio | voice content distribution system, 60 ... Reception circuit, 70 ... Information extraction apparatus, 302 ... Frame Position specifying means, 303 ... Frame extracting means, 304 ... Synchronization information adding means, 305 ... Related information adding means.

Claims (3)

Synchronization information adding means for adding synchronization information indicating a time position of the frame for each frame which is a fixed length section for audio data input;
Related information adding means for embedding related information related to the audio data in a frame of the audio data to which the synchronization information is added by the synchronization information adding means;
An audio signal processing device comprising: Predetermined processing is performed for each frame that is a fixed-length section with respect to audio data input, and synchronization information indicating the time position of the frame is added to the audio data, and the frame includes the synchronization information Information embedded with related information related to the audio data embedded in the stream data format, the synchronization information detecting means for detecting the synchronization information;
A frame extraction unit having a function of detecting a frame start position of the audio data embedded with the information based on the synchronization information detected by the synchronization information detection unit;
Related information extraction means for performing predetermined processing for each frame of the audio data embedded with the information based on the frame information detected by the frame extraction means, and extracting related information embedded in the frame When,
An audio signal processing device comprising:   As the synchronization information, a predetermined low frequency band which is a triangular wave signal having a waveform in which two adjacent frames of the audio data are synchronized with the frame as one cycle and whose amplitude changes in a triangular shape and has little influence on the sense of hearing. The audio signal processing apparatus according to claim 1, wherein the signal is used.

Images