WO2005036527A1 - スペクトル包絡線符号化のための時間境界及び周波数分解能の決定方法 - Google Patents
スペクトル包絡線符号化のための時間境界及び周波数分解能の決定方法 Download PDFInfo
- Publication number
- WO2005036527A1 WO2005036527A1 PCT/JP2004/014977 JP2004014977W WO2005036527A1 WO 2005036527 A1 WO2005036527 A1 WO 2005036527A1 JP 2004014977 W JP2004014977 W JP 2004014977W WO 2005036527 A1 WO2005036527 A1 WO 2005036527A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- boundary
- time
- time boundary
- frequency
- frequency resolution
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 92
- 238000001228 spectrum Methods 0.000 title claims description 23
- 230000001052 transient effect Effects 0.000 claims abstract description 77
- 238000011156 evaluation Methods 0.000 claims abstract description 3
- 230000003595 spectral effect Effects 0.000 claims description 18
- 230000005236 sound signal Effects 0.000 claims description 16
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000002123 temporal effect Effects 0.000 claims description 5
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000003786 synthesis reaction Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims 4
- 241000233855 Orchidaceae Species 0.000 claims 1
- 230000006870 function Effects 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 26
- 238000012545 processing Methods 0.000 description 15
- 230000011218 segmentation Effects 0.000 description 4
- 230000009897 systematic effect Effects 0.000 description 4
- 230000003321 amplification Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 101000873502 Homo sapiens S-adenosylmethionine decarboxylase proenzyme Proteins 0.000 description 1
- 241000446313 Lamella Species 0.000 description 1
- 102100035914 S-adenosylmethionine decarboxylase proenzyme Human genes 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- NUHSROFQTUXZQQ-UHFFFAOYSA-N isopentenyl diphosphate Chemical compound CC(=C)CCO[P@](O)(=O)OP(O)(O)=O NUHSROFQTUXZQQ-UHFFFAOYSA-N 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/038—Speech enhancement, e.g. noise reduction or echo cancellation using band spreading techniques
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/02—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using spectral analysis, e.g. transform vocoders or subband vocoders
- G10L19/022—Blocking, i.e. grouping of samples in time; Choice of analysis windows; Overlap factoring
- G10L19/025—Detection of transients or attacks for time/frequency resolution switching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
- H04B1/66—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission for reducing bandwidth of signals; for improving efficiency of transmission
- H04B1/667—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission for reducing bandwidth of signals; for improving efficiency of transmission using a division in frequency subbands
Definitions
- the present invention provides a systematic segment that determines time boundaries and frequency resolution for bandwidth extension techniques for sub-band coding techniques such as SBR (Spectral Band Report). About In particular, it relates to a method of dividing a time boundary and a frequency boundary and dividing each frame into segments suitable for subband coding.
- SBR Specific Band Report
- the purpose of speech coding is to convert a digitized speech stream into a compressed bit stream in a speech coding device, and as a result, even after processing the bit stream in a decoder.
- the goal is to maintain the sound quality as close to the original sound as possible.
- the well-known compression method is as shown in Fig.1.
- Figure 1 shows a typical speech coding system consisting of a decoder and an encoder.
- the module 1000 divides the audio signal in the time domain into continuous frames
- the module 10010 converts each frame of the audio signal into a frequency band
- Figure 2 is a typical time-Z frequency grid diagram for speech coding.
- module 11010 converts the audio signal into a frequency band is to use a time-Z frequency grid as shown in FIG.
- a filter bank is used to divide an audio signal into multiple subbands, and each subband represents a part of a signal within a narrow frequency range in the time domain.
- the audio frequency band is inversely quantized by the module 1030 and converted into an audio frame by the module 1040. The audio frames are then assembled in module 1505 to properly form a continuous audio stream.
- FIG. 3 shows that limiting the bandwidth at the bit rate causes high frequency lengths and loss of harmony.
- FIG. 3 shows the above-mentioned band limiting process, and reference numeral 202 denotes the result of the bandwidth of the encoded voice.
- the bandwidth extension is to recover the high frequency band by encoding the audio signal using very little additional pits in the high frequency band.
- SBR method International Publication W098 / 57436
- the bandwidth Fig. 4 is a diagram showing an example of an encoder having a sub-band encoding function for extension Fig. 4 shows a configuration of an encoder of an SBR method related to the present invention.
- the signal is banded into N subbands using the N subband filters in the 31010 analysis filter bank. Each analysis filter extracts a certain frequency range of the signal.
- the N signals generated by this filter are decimated to eliminate redundancy.
- the bandwidth extension encoder 30020 extracts some information from the filter output, and as a result, is used as information for extending the bandwidth of the audio signal of the low-frequency sub-band in the decoder. .
- the bandwidth extension information is sent to the bitstream multiplexer along with the output from the core codec 300, which encodes the low-frequency subband audio signal to form a bitstream.
- a normal SBR frame consists of a subband filter exit of L samples.
- FIG. 5 is a diagram showing a decoder having a sub-band coding function for bandwidth extension.
- FIG. 5 shows a decoder of the SBR method related to the present invention.
- the bitstream is demultiplexed to become a core audio bitstream and a bandwidth extension bitstream.
- the core decoder 4010 decodes the core audio bitstream to generate a band-limited audio signal in the time domain.
- the band-limited audio signal is
- the high frequency subband is synthesized at this subband level using the bandwidth extension information.
- the new high-frequency subband is upsampled in the same way as the low-frequency subband, and is synthesized by a synthesis filter bank 400 that outputs the final bandwidth extension signal having N filters.
- the output from the analysis filter bank 3 10 10 can be seen in the time Z frequency grid diagram of the audio signal of FIG.
- the time-frequency grid diagram is first divided into time segments shown in the time axis, and then into frequency bands shown in the frequency axis. Each frequency band has its average energy calculated, quantized, and encoded. Is encoded.
- This process is known as spectral envelope coding. That is, in spectrum envelope coding, a speech signal is represented by a distribution of average energy in each segment on a two-dimensional plane formed by a time axis and a frequency axis.
- FIG. 6 is a diagram showing segments in the time direction and the frequency direction.
- Figure 6 illustrates such a segmentation process and is shown in the IPP (W001 / 26095A1).
- 510 indicates a division in the time axis direction.
- 50 0 20 indicates division in the frequency axis direction.
- the data generated in this process is used to shape the energy of the synthesized high-frequency band, and as a result, an energy envelope similar to that of the original sound signal is obtained. Without proper segmentation, low energy—areas would be forced to share the same average energy value as high energy areas. Also, false amplification may occur at the decoder, which may lead to audibly perceptible noise.
- FIG. 7 is a diagram showing a boundary relationship between four frame types.
- the fixed boundaries 600, 670, 610 coincide with the defined SBR frames, boundaries 610, 620, 650. .
- the boundaries of the current frame, 680 and 690, are within the allowable range to erode during the following prescribed SBR frame.
- Either the start or end boundary of a variable SBR frame can be a fixed boundary or a variable boundary. If both the start and end boundaries are fixed, the variable SBR frame will match the specified SBR frame. The end boundary of the current SBR frame automatically becomes the start boundary of the next SBR frame.
- the SBR The frame is further subdivided into several time segments by intermediate boundaries. If the start and end boundaries are both fixed boundaries, the SBR frame is divided into uniform time segments. This is known in the prior art as a FIXFIX frame (ie, a FIX boundary as the start boundary and a FIX boundary as the end boundary).
- FIG. 8 is a diagram showing a FIXFIX frame using a fixed start boundary and a fixed end boundary. As shown in FIG. 8, buffer 0 10 is the start boundary and 720 0 is the end boundary. If the threshold detector detects a transient area in the current SBR frame, its end boundary is a variable boundary that must be equal to or greater than the next specified SBR frame.
- Figure 9 shows a FIXVAR frame that uses a fixed start boundary, a variable end boundary that is larger than the specified SBR frame boundary, and some intermediate boundaries specified by correlation with the end boundary, or correlation with each other. is there.
- This frame has a F IX (fixed) boundary as the starting boundary 800 1 0 and a V A R (variable) boundary as the ending boundary 800 5 0.
- the intermediate boundaries 800,0,800,800 are specified relative to each other or to the variable boundary, where d0, d1, d2, etc. are relative boundary distances.
- the first relative distance dO must start at the variable boundary. Consecutive relative distances start at the previously determined intermediate boundary.
- FIG. 4 is a diagram showing a VARVAR frame using an intermediate boundary of FIG. VAR In the VAR frame, Intermediate boundaries are relatively specified using any one of the variable boundaries.
- the middle boundary 920 is determined relative to the start boundary 910, and the middle boundaries 930, 904, 950 are mutually or variable end boundaries It is determined relatively to 9 0 6 0.
- FIG. 11 shows such a frame, a VARFIX frame that uses a variable start boundary, a fixed end boundary, and some intermediate boundaries identified by correlation with the start boundary, or correlation with each other. is there.
- 100 0 10 is a variable start boundary
- 100 0 50 is a fixed end boundary.
- 1 0 0 2 0, 1 0 0 3 0, 1 0 0 4 0 constitute dO, cH, d2 forces, etc., and gradually derived intermediate boundaries.
- the relative boundary distance between the intermediate boundary and the variable boundary can only take on a few predetermined sizes.
- FIG. 12 is a diagram showing a boundary relationship between a high-resolution time segment and a low-resolution time segment.
- Figure 12 shows the boundary relationship between two possible resolutions, a high resolution division and a low resolution division.
- the boundary of the low resolution division and the boundary of the high resolution division are mutually variable. Disclosure of the invention
- the start boundary is determined based on the end boundary of the preceding SBR frame, and the transient event boundary is determined using the threshold detector.
- some method is needed to determine the end boundary and all intermediate boundaries.
- a device is needed to determine the end boundary and all intermediate boundaries.
- the spectrum coding method used in the conventional technology has low time resolution but high spectrum resolution (for example, frequency spectrum is derived over many frequency bands for a few time segments). Do).
- the prior art uses a high temporal resolution but a low spectral resolution (eg, for many time segments) to encode the band after the transient. Derive the frequency spectrum of a smaller frequency band).
- the level of resolution is changed because transient events are more likely to show temporal changes than spectral changes.
- Lower frequency resolution can prevent a sharp increase in bit consumption.
- this method shows that later transient bands show large spectral changes that require more sophisticated resolution, for example, indicating a sudden onset of a tone signal. In some cases it is not enough.
- the present invention provides a systematic method for determining the end boundary and all intermediate boundaries, taking into account all the standard constraints imposed by the decoder. The purpose is to:
- the frame type of the current SBR frame is determined depending on the type of the end boundary of the preceding frame and the presence of a transient in the current SBR frame.
- the start boundary is also determined according to the end boundary of the preceding SBR frame.
- a setting with a low time resolution is used. That is, as shown in FIG. 8, the intermediate boundaries are set at equal intervals of the time width d.
- the search for possible intermediate boundaries is performed first in the region after the transient time position.
- the end boundary is also determined at this stage. Then, if the maximum number of boundaries within the allowable range has not been exhausted in the first stage, a search is made to determine whether an intermediate boundary can be set in the region temporally before the position of the transient event.
- one search must be performed in all regions bounded by a variable start boundary and a fixed end boundary. All of the above is achieved by two forward search operations and one backward search operation. They use the same principle. It is based on assessing signal changes in the time segment, but slightly different to fit the scenario being adapted.
- the present invention provides an adaptive method for objectively assessing energy changes in the spectral direction.
- the boundaries of the low resolution section are alternative boundaries of the high resolution section, high resolution is assumed first and the average energy is calculated in each frequency band. Energy ratios are calculated for all pairs of frequency bands bounded by the low resolution boundary. If the minimum value of all energy differences calculated in all time intervals exceeds a predetermined threshold, high frequency resolution is adopted. Otherwise, low frequency resolution is used. Given the importance of providing high temporal resolution in the region of later transients, the method employs strict criteria for applying high frequency resolution in this region.
- the method for determining the time boundary and the frequency resolution of the spectrum envelope coding is a method for determining the time boundary and the frequency resolution in the spectrum envelope coding of the audio signal using the time-frequency grid. Then, the start time boundary of the current frame is derived from the end time boundary of the preceding frame in the envelope data, and the transition time is calculated from the spectrum data between the start time boundary and the end time boundary within a predetermined allowable range.
- the event detector detects the time position of a transient event whose magnitude is greater than or equal to a certain value, and compares the magnitude of the transient event with a predetermined signal change criterion to determine the current frame within the allowable range.
- the time boundary determination method provides a preferred systematic method of performing frame segmentation of the transient region before and after by evaluating the change in energy in the time direction. This method takes into account all the imposed standard constraints, giving priority to the area behind the transient, or to the area farther from the transient. Good sound quality is realized by giving priority to a region that is closer. Further, the frequency resolution determining method applied in the present invention detects an energy distribution in the frequency direction in a region behind the transient event.
- the energy of all frequency bands bordering on a low-resolution boundary indicated by a predetermined frequency is evaluated, whereby the time segment is evaluated. Derive the frequency resolution in the unit.
- This method uses high-resolution partitioning when large changes in the energy distribution are found.
- the present invention can be realized not only as a method for determining the time boundary and frequency resolution of such spectral envelope coding, but also as a program for causing a computer to execute those steps. You can also. Further, these steps can be realized as an audio encoding device and an audio decoding device incorporated in an integrated circuit or the like. It goes without saying that such a program can be distributed via a recording medium such as CD-ROM or a transmission medium such as the Internet.
- FIG. 1 illustrates a typical speech coding system.
- FIG. 2 is a typical time Z frequency grid diagram in speech coding.
- Figure 3 shows that limiting the bandwidth in the bit rate causes high frequency lengths and loss of harmony.
- Figure 4 shows an encoder with a subband encoding function for bandwidth expansion. It is a figure showing the example of.
- FIG. 5 is a diagram illustrating a decoder having a subband encoding function for bandwidth extension.
- FIG. 6 is a diagram showing segments in the time direction and the frequency direction.
- FIG. 7 is a diagram showing a boundary relationship between four frame types.
- FIG. 8 is a diagram showing a FIXFIX frame using a fixed start boundary and a fixed end boundary.
- Figure 9 shows a FIXVAR frame that uses a fixed start boundary, a variable end boundary that is larger than the specified SBR frame boundary, and some intermediate boundaries specified by correlation with the end boundary, or correlation with each other. is there.
- Figure 10 shows a VARVAR frame with a variable start boundary, a variable end boundary that is larger than the defined SBR frame boundary, and some intermediate boundaries identified by correlation with two end boundaries, or correlation with each other. The figure shows
- FIG. 11 shows a VARFIX frame using a variable start boundary, a fixed end boundary, and some intermediate boundaries specified by correlation with the start boundary, or with each other.
- FIG. 12 is a diagram showing a boundary relationship between a high-resolution time segment and a low-resolution time segment.
- FIG. 13 is a diagram showing a flowchart illustrating the entirety of the time boundary determination unit according to the present invention.
- FIG. 14 is a diagram illustrating the use of three search types in the variable part of four frame types.
- FIG. 15 is a diagram showing a flowchart of the forward search (type 1) process.
- Figure 16 is a diagram showing a flowchart of the forward search (type 2) process.
- FIG. 17 is a diagram showing a flowchart of the backward search processing.
- FIG. 18 is an illustration showing the frequency resolution determining section according to the present invention. BEST MODE FOR CARRYING OUT THE INVENTION
- FIG. 13 is a diagram showing a flowchart illustrating the entire time boundary determining unit according to the present invention.
- FIG. 13 shows an overall view of the processing for determining all time boundaries.
- the first border [0] is set at the end border of the preceding SBR frame, and noBorder of the border counter is initialized to 1.
- the transient detector is activated in the current frame. This is to detect the most abrupt transient behavior from borderO to (SBR boundary + V in the next rule). This detects the presence or absence of a transient in the current frame. Where V is the amount that goes into the next SBR frame allowed by the configuration.
- the type of end boundary of the preceding SBR frame is checked at 12030. If it is a fixed boundary, the current frame will be of the FIXVAR type in 1250. If it is a variable border, the current frame will be of type VARVAR in "! 290.90. In either case, the border indicating the location of the transient will be in border 1 The noBorder registered and indicating the number of borders in the current frame is incremented and incremented.
- the type of the end boundary of the preceding SBR frame is checked at 12040. If it is a fixed boundary, the current frame will be of type FIXFIX in 121,300. If it is a variable boundary, the current frame will be of VARFIX type during 1250.
- the area between the transient and should be checked to determine the need for an intermediate boundary.
- the method of forward search (type 1) described in 3.5.1.2 is used for this purpose.
- the number of borders noBorder is confirmed at 12070. If noBorder is less than the maximum value MaxBorder allowed for the number of borders, then use backward search at 12080 to identify the area between the transient and the starting border. If, instantiate more intermediate boundaries.
- the above-described sequence processing gives priority to an intermediate boundary detected in a region after the transient event position.
- the forward search (type 1) described in 3.5.1.2, at 1200, due to the potential need for intermediate boundaries Using the method described above, confirm the area between the transient event and (SBR boundary + V specified below). At the end of the forward search, the noBorder force is confirmed during 1 2 1 1 1 0. If the noBorder force is less than the maximum value of the number of boundaries MaxBorder, another forward search is performed at 1 2 1 2 0 to identify the region between the transient; Use type 2) and instantiate more intermediate boundaries if necessary. Same as above, but the above sequence processing Gives priority to intermediate boundaries detected in the area behind the transient event location. If the current frame is FIXFIX, select a low time resolution setting in 1240. More details are described in 3.5-2.
- the forward search (type 1) method described above is used for this purpose.
- the four branches of each process described above are finished by rearranging the boundaries detected for subsequent processes in ascending order in 1270.
- FIG. 14 illustrates the use of three search types in the variable part of four frame types.
- Figure 14 depicts the use of three search types among the four frame types, where 1 0 1 0 and 1 0 0 0 indicate the forward search process (type 1) and 1 740 and 17050 indicate forward search processing (type 2), and 17030 indicates backward search processing.
- the region after the transient event position is prioritized in the intermediate boundary determination process in the above embodiment, but it is also possible to select which region is prioritized by evaluating the signal change It becomes. If the signal change is larger in the region before the transient event, the region before the transient event has priority, and the signal change is larger in the region after the transient event. In some cases, the area after the transient takes precedence.
- This forward search method (Type 1) is designed for regions that start at the location of the transient and end at a variable boundary that has not yet been determined. Its purpose is to define intermediate and end boundaries.
- Three input Z ⁇ The lamellas border1, border2, noBorderLimit are initialized according to 1206 and 1210 in Figure 13 to set the search area (between border1 and border2) It must be.
- NoBorderLimit is initialized to the maximum number of boundaries allowed.
- FIG. 15 is a flowchart showing a forward search (type 1) process.
- Figure 15 shows a flowchart of this method.
- This method uses two intermediate variables, i and j, to represent the left and right boundaries of the time segment. Let the left boundary of the segment be i and the right boundary be j. k is used to refer to the relative boundary distance D k of the current time segment. This is because all relative distances can take only predetermined sizes (di [D1, D2, D3, D4], with 0 ⁇ D1 ⁇ D2 ⁇ D3 ⁇ D4.).
- ⁇ is still below the specified SBR frame boundary (i.e., the time position of the transient does not exceed the specified SBR frame boundary to be the current frame) Is determined.
- D k is not yet D 4
- 1 3 5 0 is no and the variable peak_ratio shown below is in 1 3 0 6 0 due to the region between i and (j-1) Be evaluated.
- the criterion for determining a new intermediate boundary is based on identifying the average energy of all time segments and the energy-ratio of each time position. It is performed in 13070 by the following equation.
- ET m is the energy at time position m.
- E T is the average energy of all time positions calculated from i to (j-1).
- D r 1 is a predetermined threshold.
- Another possible signal change criterion may be based on comparing the following maximum and minimum energy values:
- the signal change criteria may be based on comparing the maximum value and the minimum value of the absolute amplitude below.
- the signal change is considered to be sufficiently quiet, so that 1 3 0 8 0 In which the larger time span D k is applied by first increasing the value of k and then adjusting j. That is, boundaries are created at larger time intervals where signal changes are small.
- i is initialized to refer to the ending boundary.
- the relative boundary distance between borde r i and bor de r (i-1) is subsequently confirmed. If the relative boundary distance is not smaller than D4, the relative boundary distance cannot be expanded and i is reduced, so that bor de r (i-1) and bor de r (i- 2) The relative boundary distance between is confirmed. However, if the relative boundary distance is less than D4, the relative distance between bor der i and bor der (i-1) is extended in 1316. This process is repeated until the end boundary is greater than or equal to the specified SBR frame boundary, as verified during 13170.
- This forward search method (type 2) is set in a band that starts with a variable or fixed boundary and ends with a previously determined boundary. Unlike the Type 1 forward search method, the objective is only to determine the intermediate boundaries. The three input parameters, border1, border2, and noBor derL imit, are used to set the maximum number of boundaries within the search area and allowable range. Must be initialized in accordance with.
- FIG. 16 is a flowchart showing a forward search (type 2) process. Figure 16 shows a flowchart of this method. In principle, the two forward search methods are the same. Therefore, the processing from 1 410 to 1 410 is almost the same as the processing from 1 310 to 1 310 in Figure 15 with some exceptions. .
- the peak_ratio of the new boundary must be preserved when it is initialized. This is redundant in 1 4 1 1 0 This is to remove the border. These redundant boundaries may be created to set the time segment length in the current frame to the maximum allowed. This is because the position of each boundary is represented by a relative position with respect to each other, so that it may be necessary to set a redundant boundary for the sake of convenience in order to set the next boundary. But if this is an end boundary, it does not cause a problem.
- This backward search method is set for the region of the boundary that starts with the transient and the boundary that ends with the start boundary.
- the three input parameters, borderK border2, and noBorderL imit, must be initialized according to the method of FIG. 13 at 12080 to set the search band and the maximum number within the allowed range.
- FIG. 17 is a diagram showing a flowchart of the backward search process.
- this method is similar to forward search (type 2). Therefore, except for the processing executed in the reverse direction, the processing from 15010 to 151110 is almost the same as the processing from 14010 to 14110.
- the time domain diagram for the FIX FIX frame is uniformly partitioned by intermediate boundaries based on the number of selected boundaries.
- a simple way to select the number of boundaries is to try using the lowest number of boundaries and evaluate the peak-ratio of the formed time segment. If any of the peak_ratio exceeds a certain threshold, a larger number of boundaries is required and the evaluation of the peak_ratio for each time segment formed is repeated. The process ends when the peak_ratio of the entire time segment formed falls below a certain threshold or when the maximum number of boundaries is reached.
- FIG. 18 is an illustration showing the frequency division determining section according to the present invention.
- the boundary of the low resolution division is the boundary of every other high resolution division.
- the average energy of all frequency bands in the time segment is calculated, assuming that high frequency resolution is applied.
- the average energy is shown using E i.
- the high frequency resolution is selected by satisfying the following equation. If not. Low frequency resolution is selected. 1, 2 ...
- the average energy is used for the determination, but other parameters such as amplitude information indicating the signal change could be used instead.
- the encoding device of the present invention is useful as an audio signal encoding device provided in a personal computer having a communication function, a PDA digital broadcasting station, a distribution server for distributing contents, a mobile phone, and the like.
- the decoding device according to the present invention is adapted to receive distribution of music content or the like. It is useful as a personal computer with a communication function for reading data and a recording medium reading function, a PDA, a mobile phone, a portable audio reproduction device, and an audio decoding device provided in an STB for receiving digital broadcasts.
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Quality & Reliability (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/567,601 US7451091B2 (en) | 2003-10-07 | 2004-10-04 | Method for determining time borders and frequency resolutions for spectral envelope coding |
DE602004030594T DE602004030594D1 (de) | 2003-10-07 | 2004-10-04 | Verfahren zur entscheidung der zeitgrenze zur codierung der spektro-hülle und frequenzauflösung |
EP04773711A EP1672618B1 (en) | 2003-10-07 | 2004-10-04 | Method for deciding time boundary for encoding spectrum envelope and frequency resolution |
JP2005514625A JP4767687B2 (ja) | 2003-10-07 | 2004-10-04 | スペクトル包絡線符号化のための時間境界及び周波数分解能の決定方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2003-348621 | 2003-10-07 | ||
JP2003348621 | 2003-10-07 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2005036527A1 true WO2005036527A1 (ja) | 2005-04-21 |
Family
ID=34430971
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2004/014977 WO2005036527A1 (ja) | 2003-10-07 | 2004-10-04 | スペクトル包絡線符号化のための時間境界及び周波数分解能の決定方法 |
Country Status (5)
Country | Link |
---|---|
US (1) | US7451091B2 (ja) |
EP (1) | EP1672618B1 (ja) |
JP (1) | JP4767687B2 (ja) |
DE (1) | DE602004030594D1 (ja) |
WO (1) | WO2005036527A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008129542A (ja) * | 2006-11-24 | 2008-06-05 | Fujitsu Ltd | 復号化装置および復号化方法 |
JP2008129541A (ja) * | 2006-11-24 | 2008-06-05 | Fujitsu Ltd | 復号化装置および復号化方法 |
JP2008224902A (ja) * | 2007-03-09 | 2008-09-25 | Fujitsu Ltd | 符号化装置および符号化方法 |
JP2013148920A (ja) * | 2009-01-16 | 2013-08-01 | Dolby International Ab | クロス生成物により向上された高調波転換 |
Families Citing this family (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1761916A1 (en) * | 2004-06-22 | 2007-03-14 | Koninklijke Philips Electronics N.V. | Audio encoding and decoding |
US7720230B2 (en) * | 2004-10-20 | 2010-05-18 | Agere Systems, Inc. | Individual channel shaping for BCC schemes and the like |
US20060269024A1 (en) * | 2005-05-27 | 2006-11-30 | Francis Dominique | Initial multi-path acquisition of random access channels |
US7764656B2 (en) * | 2005-07-13 | 2010-07-27 | Alcatel-Lucent Usa Inc. | Methods of multipath acquisition for dedicated traffic channels |
US7929499B2 (en) * | 2005-07-13 | 2011-04-19 | Alcatel-Lucent Usa Inc. | Methods of multipath acquisition for dedicated traffic channels |
US7856071B2 (en) * | 2005-07-26 | 2010-12-21 | Alcatel-Lucent Usa Inc. | Multi-path acquisition in the presence of very high data rate users |
DE102006049154B4 (de) * | 2006-10-18 | 2009-07-09 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Kodierung eines Informationssignals |
US8126721B2 (en) | 2006-10-18 | 2012-02-28 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Encoding an information signal |
US8417532B2 (en) | 2006-10-18 | 2013-04-09 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Encoding an information signal |
US8041578B2 (en) * | 2006-10-18 | 2011-10-18 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Encoding an information signal |
US8428953B2 (en) * | 2007-05-24 | 2013-04-23 | Panasonic Corporation | Audio decoding device, audio decoding method, program, and integrated circuit |
EP2242048B1 (en) * | 2008-01-09 | 2017-06-14 | LG Electronics Inc. | Method and apparatus for identifying frame type |
CN102089814B (zh) | 2008-07-11 | 2012-11-21 | 弗劳恩霍夫应用研究促进协会 | 对编码的音频信号进行解码的设备和方法 |
AU2009267532B2 (en) * | 2008-07-11 | 2013-04-04 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | An apparatus and a method for calculating a number of spectral envelopes |
CN101751926B (zh) * | 2008-12-10 | 2012-07-04 | 华为技术有限公司 | 信号编码、解码方法及装置、编解码*** |
ES2400661T3 (es) * | 2009-06-29 | 2013-04-11 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Codificación y decodificación de extensión de ancho de banda |
AU2014201331B2 (en) * | 2009-06-29 | 2015-10-01 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Bandwidth extension encoder, bandwidth extension decoder and phase vocoder |
MY156027A (en) | 2010-08-12 | 2015-12-31 | Fraunhofer Ges Forschung | Resampling output signals of qmf based audio codecs |
JP5633431B2 (ja) * | 2011-03-02 | 2014-12-03 | 富士通株式会社 | オーディオ符号化装置、オーディオ符号化方法及びオーディオ符号化用コンピュータプログラム |
KR101572034B1 (ko) | 2011-05-19 | 2015-11-26 | 돌비 레버러토리즈 라이쎈싱 코오포레이션 | 파라메트릭 오디오 코딩 방식들의 포렌식 검출 |
CN104143341B (zh) * | 2013-05-23 | 2015-10-21 | 腾讯科技(深圳)有限公司 | 爆音检测方法和装置 |
FR3008533A1 (fr) * | 2013-07-12 | 2015-01-16 | Orange | Facteur d'echelle optimise pour l'extension de bande de frequence dans un decodeur de signaux audiofrequences |
JP6763194B2 (ja) * | 2016-05-10 | 2020-09-30 | 株式会社Jvcケンウッド | 符号化装置、復号装置、通信システム |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0685607A (ja) * | 1992-08-31 | 1994-03-25 | Alpine Electron Inc | 高域成分復元装置 |
JPH06214600A (ja) * | 1992-12-14 | 1994-08-05 | American Teleph & Telegr Co <Att> | 汎用合成による分析符号化の時間軸シフト方法とその装置 |
JPH11504733A (ja) * | 1996-02-26 | 1999-04-27 | エイ・ティ・アンド・ティ・コーポレーション | 聴覚モデルによる量子化を伴う予測残余信号の変形符号化による多段音声符号器 |
JP2000267700A (ja) * | 1999-03-17 | 2000-09-29 | Yrp Kokino Idotai Tsushin Kenkyusho:Kk | 音声符号化復号方法および装置 |
JP2003514267A (ja) * | 1999-11-18 | 2003-04-15 | ボイスエイジ コーポレイション | 広帯域音声及びオーディオ信号復号器における利得平滑化 |
JP2004053940A (ja) * | 2002-07-19 | 2004-02-19 | Matsushita Electric Ind Co Ltd | オーディオ復号化装置およびオーディオ復号化方法 |
JP2004102095A (ja) * | 2002-09-12 | 2004-04-02 | Sony Corp | 信号処理システム、信号処理装置および方法、記録媒体、並びにプログラム |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5956674A (en) | 1995-12-01 | 1999-09-21 | Digital Theater Systems, Inc. | Multi-channel predictive subband audio coder using psychoacoustic adaptive bit allocation in frequency, time and over the multiple channels |
JP3418305B2 (ja) * | 1996-03-19 | 2003-06-23 | ルーセント テクノロジーズ インコーポレーテッド | オーディオ信号を符号化する方法および装置および知覚的に符号化されたオーディオ信号を処理する装置 |
SE512719C2 (sv) * | 1997-06-10 | 2000-05-02 | Lars Gustaf Liljeryd | En metod och anordning för reduktion av dataflöde baserad på harmonisk bandbreddsexpansion |
US6377915B1 (en) | 1999-03-17 | 2002-04-23 | Yrp Advanced Mobile Communication Systems Research Laboratories Co., Ltd. | Speech decoding using mix ratio table |
US6978236B1 (en) * | 1999-10-01 | 2005-12-20 | Coding Technologies Ab | Efficient spectral envelope coding using variable time/frequency resolution and time/frequency switching |
US7110953B1 (en) * | 2000-06-02 | 2006-09-19 | Agere Systems Inc. | Perceptual coding of audio signals using separated irrelevancy reduction and redundancy reduction |
SE0301273D0 (sv) * | 2003-04-30 | 2003-04-30 | Coding Technologies Sweden Ab | Advanced processing based on a complex-exponential-modulated filterbank and adaptive time signalling methods |
-
2004
- 2004-10-04 US US10/567,601 patent/US7451091B2/en not_active Expired - Fee Related
- 2004-10-04 JP JP2005514625A patent/JP4767687B2/ja not_active Expired - Fee Related
- 2004-10-04 DE DE602004030594T patent/DE602004030594D1/de active Active
- 2004-10-04 EP EP04773711A patent/EP1672618B1/en not_active Expired - Fee Related
- 2004-10-04 WO PCT/JP2004/014977 patent/WO2005036527A1/ja active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0685607A (ja) * | 1992-08-31 | 1994-03-25 | Alpine Electron Inc | 高域成分復元装置 |
JPH06214600A (ja) * | 1992-12-14 | 1994-08-05 | American Teleph & Telegr Co <Att> | 汎用合成による分析符号化の時間軸シフト方法とその装置 |
JPH11504733A (ja) * | 1996-02-26 | 1999-04-27 | エイ・ティ・アンド・ティ・コーポレーション | 聴覚モデルによる量子化を伴う予測残余信号の変形符号化による多段音声符号器 |
JP2000267700A (ja) * | 1999-03-17 | 2000-09-29 | Yrp Kokino Idotai Tsushin Kenkyusho:Kk | 音声符号化復号方法および装置 |
JP2003514267A (ja) * | 1999-11-18 | 2003-04-15 | ボイスエイジ コーポレイション | 広帯域音声及びオーディオ信号復号器における利得平滑化 |
JP2004053940A (ja) * | 2002-07-19 | 2004-02-19 | Matsushita Electric Ind Co Ltd | オーディオ復号化装置およびオーディオ復号化方法 |
JP2004102095A (ja) * | 2002-09-12 | 2004-04-02 | Sony Corp | 信号処理システム、信号処理装置および方法、記録媒体、並びにプログラム |
Non-Patent Citations (1)
Title |
---|
See also references of EP1672618A4 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008129542A (ja) * | 2006-11-24 | 2008-06-05 | Fujitsu Ltd | 復号化装置および復号化方法 |
JP2008129541A (ja) * | 2006-11-24 | 2008-06-05 | Fujitsu Ltd | 復号化装置および復号化方法 |
US8249882B2 (en) | 2006-11-24 | 2012-08-21 | Fujitsu Limited | Decoding apparatus and decoding method |
JP2008224902A (ja) * | 2007-03-09 | 2008-09-25 | Fujitsu Ltd | 符号化装置および符号化方法 |
US8073050B2 (en) | 2007-03-09 | 2011-12-06 | Fujitsu Limited | Encoding device and encoding method |
JP2013148920A (ja) * | 2009-01-16 | 2013-08-01 | Dolby International Ab | クロス生成物により向上された高調波転換 |
US9799346B2 (en) | 2009-01-16 | 2017-10-24 | Dolby International Ab | Cross product enhanced harmonic transposition |
US10192565B2 (en) | 2009-01-16 | 2019-01-29 | Dolby International Ab | Cross product enhanced harmonic transposition |
US10586550B2 (en) | 2009-01-16 | 2020-03-10 | Dolby International Ab | Cross product enhanced harmonic transposition |
US11031025B2 (en) | 2009-01-16 | 2021-06-08 | Dolby International Ab | Cross product enhanced harmonic transposition |
US11682410B2 (en) | 2009-01-16 | 2023-06-20 | Dolby International Ab | Cross product enhanced harmonic transposition |
US11935551B2 (en) | 2009-01-16 | 2024-03-19 | Dolby International Ab | Cross product enhanced harmonic transposition |
Also Published As
Publication number | Publication date |
---|---|
US7451091B2 (en) | 2008-11-11 |
DE602004030594D1 (de) | 2011-01-27 |
JP4767687B2 (ja) | 2011-09-07 |
JPWO2005036527A1 (ja) | 2006-12-21 |
EP1672618B1 (en) | 2010-12-15 |
EP1672618A4 (en) | 2008-06-25 |
EP1672618A1 (en) | 2006-06-21 |
US20060256971A1 (en) | 2006-11-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2005036527A1 (ja) | スペクトル包絡線符号化のための時間境界及び周波数分解能の決定方法 | |
JP4918841B2 (ja) | 符号化システム | |
CN101512639B (zh) | 用于语音/音频发送器和接收器的方法和设备 | |
KR100348368B1 (ko) | 디지털 음향 신호 부호화 장치, 디지털 음향 신호 부호화방법 및 디지털 음향 신호 부호화 프로그램을 기록한 매체 | |
US8856049B2 (en) | Audio signal classification by shape parameter estimation for a plurality of audio signal samples | |
US8112286B2 (en) | Stereo encoding device, and stereo signal predicting method | |
US9842603B2 (en) | Encoding device and encoding method, decoding device and decoding method, and program | |
EP1440300B1 (en) | Encoding device, decoding device and audio data distribution system | |
JP5942358B2 (ja) | 符号化装置および方法、復号装置および方法、並びにプログラム | |
US9779749B2 (en) | Audio signal coding method and apparatus | |
JP3739959B2 (ja) | デジタル音響信号符号化装置、デジタル音響信号符号化方法及びデジタル音響信号符号化プログラムを記録した媒体 | |
KR101967122B1 (ko) | 신호 처리 장치 및 방법, 및 프로그램 | |
US20040186735A1 (en) | Encoder programmed to add a data payload to a compressed digital audio frame | |
WO2004038927A1 (en) | Packet loss recovery based on music signal classification and mixing | |
EP1441330B1 (en) | Method of encoding and/or decoding digital audio using time-frequency correlation and apparatus performing the method | |
CN111968655A (zh) | 信号编码方法和装置以及信号解码方法和装置 | |
EP4143825A2 (en) | Method, apparatus and system for enhancing multi-channel audio in a dynamic range reduced domain | |
JP4317355B2 (ja) | 符号化装置、符号化方法、復号化装置、復号化方法および音響データ配信システム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 2005514625 Country of ref document: JP |
|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2004773711 Country of ref document: EP Ref document number: 2006256971 Country of ref document: US Ref document number: 10567601 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 2004773711 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 10567601 Country of ref document: US |