iteh STANDARD PREVIEW (standards.iteh.ai)


 Edin Demšar
 pred 15 dnevi
 Pregledov:
Transkripcija
1 SLOVENSKI STANDARD 01december2003 Digital cellular telecommunications system (Phase 2+) (GSM); Adaptive MultiRate (AMR) speech transcoding Ta slovenski standard je istoveten z: EN Version ICS: Globalni sistem za mobilno telekomunikacijo (GSM) Global System for Mobile Communication (GSM) en Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
2
3 European Standard (Telecommunications series) Digital cellular telecommunications system (Phase 2+); Adaptive MultiRate (AMR) speech transcoding GLOBAL SYSTEM FOR MOBILE COMMUNICATIONS R
4 2 Reference DEN/SMG Q7 Keywords Digital cellular telecommunications system, Global System for Mobile communication (GSM), speech, AMR Postal address F Sophia Antipolis Cedex  FRANCE Office address 650 Route des Lucioles  Sophia Antipolis Valbonne  FRANCE Tel.: Fax: Siret SIST N 348 EN V7.1.1: NAF 742 C Association à but non lucratif enregistrée à la SousPréfecture de Grasse (06) N 7803/88 Internet Individual copies of this deliverable can be downloaded from If you find errors in the present document, send your comment to: Important notice This deliverable may be made available in more than one electronic version or in print. In any case of existing or perceived difference in contents between such versions, the reference version is the Portable Document Format (PDF). In case of dispute, the reference shall be the printing on printers of the PDF version kept on a specific network drive within Secretariat. Copyright Notification No part may be reproduced except as authorized by written permission. The copyright and the foregoing restriction extend to reproduction in all media. European Telecommunications Standards Institute All rights reserved.
5 3 Contents Intellectual Property Rights...5 Foreword Scope References Definitions, symbols and abbreviations Definitions Symbols Abbreviations Outline description Functional description of audio parts Preparation of speech samples PCM format conversion Principles of the GSM adaptive multirate speech encoder Principles of the GSM adaptive multirate speech decoder Sequence and subjective importance of encoded parameters Functional description of the encoder Preprocessing (all modes) Linear prediction analysis and quantization Windowing anditeh autocorrelation STANDARD computation...18 PREVIEW LevinsonDurbin algorithm (all modes) LP to LSP conversion (all modes) LSP to LP conversion (all modes) Quantization of the LSP coefficients SIST EN V7.1.1: Interpolation of the LSPs Monitoring resonance in the LPC spectrum (all modes) Openloop pitch analysis Impulse response computation (all modes) Target signal computation (all modes) Adaptive codebook Adaptive codebook search Adaptive codebook gain control (all modes) Algebraic codebook Algebraic codebook structure Algebraic codebook search Quantization of the adaptive and fixed codebook gains Adaptive codebook gain limitation in quantization Quantization of codebook gains Update past quantized adaptive codebook gain buffer (all modes) Memory update (all modes) Functional description of the decoder Decoding and speech synthesis Postprocessing Adaptive postfiltering (all modes) Highpass filtering and upscaling (all modes) Detailed bit allocation of the adaptive multirate codec Homing sequences Functional description Definitions Encoder homing Decoder homing...52
6 4 9 Bibliography...56 Annex A (informative): Document change history...57 History...58
7 5 Intellectual Property Rights IPRs essential or potentially essential to the present document may have been declared to. The information pertaining to these essential IPRs, if any, is publicly available for members and nonmembers, and can be found in SR : "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to in respect of standards", which is available from the Secretariat. Latest updates are available on the Web server ( Pursuant to the IPR Policy, no investigation, including IPR searches, has been carried out by. No guarantee can be given as to the existence of other IPRs not referenced in SR (or the updates on the Web server) which are, or may be, or may become, essential to the present document. Foreword This European Standard (Telecommunications series) has been produced by Special Mobile Group (SMG). The present document describes the detailed mapping from input blocks of 160 speech samples in 13bit uniform PCM format to encoded blocks of 95, 103, 118, 134, 148, 159, 204, and 244 bits and from encoded blocks of 95, 103, 118, 134, 148, 159, 204, and 244 bits to output blocks of 160 reconstructed speech samples within the digital cellular telecommunications system. The contents of the present document is subject to continuing work within SMG and may change following formal SMG approval. Should SMG modify the contents of the present document it will be rereleased with an identifying change of release date and an increase in version number as follows: Version 7.x.y where: 7 indicates Release 1998 of GSM Phase 2+ x the second digit is incremented for all changes of substance, i.e. technical enhancements, corrections, updates, etc. y the third digit is incremented when editorial only changes have been incorporated in the specification. National transposition dates Date of adoption of this EN: 3 December 1999 Date of latest announcement of this EN (doa): 31 March 2000 Date of latest publication of new National Standard or endorsement of this EN (dop/e): 30 September 2000 Date of withdrawal of any conflicting National Standard (dow): 30 September 2000
8 6 1 Scope The present document describes the detailed mapping from input blocks of 160 speech samples in 13bit uniform PCM format to encoded blocks of 95, 103, 118, 134, 148, 159, 204, and 244 bits and from encoded blocks of 95, 103, 118, 134, 148, 159, 204, and 244 bits to output blocks of 160 reconstructed speech samples. The sampling rate is samples/s leading to a bit rate for the encoded bit stream of 4.75, 5.15, 5.90, 6.70, 7.40, 7.95, 10.2 or 12.2 kbit/s. The coding scheme for the multirate coding modes is the socalled Algebraic Code Excited Linear Prediction Coder, hereafter referred to as ACELP. The multirate ACELP coder is referred to as MRACELP. In the case of discrepancy between the requirements described in the present document and the fixed point computational description (ANSIC code) of these requirements contained in GSM [6], the description in GSM [6] will prevail. The ANSIC code is not described in the present document, see GSM [6] for a description of the ANSIC code. The transcoding procedure specified in the present document is applicable for the adaptive multirate full rate and half rate speech traffic channels (TCH) in the GSM system. In GSM [5], a reference configuration for the speech transmission chain of the GSM adaptive multirate (AMR) system is shown. According to this reference configuration, the speech encoder takes its input as a 13bit uniform PCM signal either from the audio part of the Mobile Station or on the network side, from the PSTN via an 8bit Alaw or µ  law to 13bit uniform PCM conversion. The encoded speech at the output of the speech encoder is delivered to a channel encoder unit which is specified in GSM [3]. In the receive direction, the inverse operations take place. 2 References The following documents contain provisions which, through reference in this text, constitute provisions of the present document. References are either specific (identified by date of publication, edition number, version number, etc.) or nonspecific. For a specific reference, subsequent revisions do not apply. For a nonspecific reference, the latest version applies. A nonspecific reference to an ETS shall also be taken to refer to later versions published as an EN with the same number. For this Release 1998 document, references to GSM documents are for Release 1998 versions (version 7.x.y). [1] GSM 01.04: "Digital cellular telecommunications system (Phase 2+); Abbreviations and acronyms". [2] GSM 03.50: "Digital cellular telecommunications system (Phase 2+); Transmission planning aspects of the speech service in the GSM Public Land Mobile Network (PLMN) system". [3] GSM 05.03: "Digital cellular telecommunications system (Phase 2+); Channel coding". [4] GSM 06.94: "Digital cellular telecommunications system (Phase 2+); Voice Activity Detection (VAD) for Adaptive MultiRate speech traffic channels". [5] GSM 06.71: "Digital cellular telecommunications system (Phase 2+); Adaptive MultiRate speech processing functions; General description". [6] GSM 06.73: "Digital cellular telecommunications system (Phase 2+); ANSIC code for the Adaptive MultiRate speech codec".
9 7 [7] GSM 06.74: "Digital cellular telecommunications system (Phase 2+); Test sequences for the GSM Adaptive MultiRate speech codec". [8] ITUT Recommendation G.711 (1988): "Coding of analogue signals by pulse code modulation Pulse code modulation (PCM) of voice frequencies". [9] ITUT Recommendation G.726: "40, 32, 24, 16 kbit/s adaptive differential pulse code modulation (ADPCM)". 3 Definitions, symbols and abbreviations 3.1 Definitions For the purposes of the present document, the following terms and definitions apply: adaptive codebook: The adaptive codebook contains excitation vectors that are adapted for every subframe. The adaptive codebook is derived from the longterm filter state. The lag value can be viewed as an index into the adaptive codebook. adaptive postfilter: This filter is applied to the output of the shortterm synthesis filter to enhance the perceptual quality of the reconstructed speech. In the adaptive multirate codec, the adaptive postfilter is a cascade of two filters: a formant postfilter and a tilt compensation filter. Adaptive MultiRate (AMR) codec: Speech and channel codec capable of operating at gross bitrates of 11.4 kbit/s ( halfrate ) and 22.8 kbit/s ( fullrate ). In addition, the codec may operate at various combinations of speech and channel coding (codec mode) bitrates for each channel mode. algebraic codebook: A fixed codebook where algebraic code is used to populate the excitation vectors (innovation vectors). The excitation contains a small number SIST ofen nonzero pulses V7.1.1:2003 with predefined interlaced sets of positions. AMR handover: Handover between the FR and HR channel modes to optimise AMR operation. antisparseness processing: An adaptive postprocessing procedure applied to the fixed codebook vector in order to reduce perceptual artifacts from a sparse fixed codebook vector. channel mode: Halfrate or fullrate operation. channel mode adaptation: The control and selection of the (FR or HR) channel mode. channel repacking: Repacking of HR (and FR) radio channels of a given radio cell to achieve higher capacity within the cell. closedloop pitch analysis: This is the adaptive codebook search, i.e., a process of estimating the pitch (lag) value from the weighted input speech and the long term filter state. In the closedloop search, the lag is searched using error minimization loop (analysisbysynthesis). In the adaptive multirate codec, closedloop pitch search is performed for every subframe. codec mode: For a given channel mode, the bit partitioning between the speech and channel codecs. codec mode adaptation: The control and selection of the codec mode bitrates. Normally, implies no change to the channel mode. direct form coefficients: One of the formats for storing the short term filter parameters. In the adaptive multirate codec, all filters which are used to modify speech samples use direct form coefficients. fixed codebook: The fixed codebook contains excitation vectors for speech synthesis filters. The contents of the codebook are nonadaptive (i.e., fixed). In the adaptive multirate codec, the fixed codebook is implemented using an algebraic codebook.
10 8 fractional lags: A set of lag values having subsample resolution. In the adaptive multirate codec a subsample resolution of 1/6th or 1/3rd of a sample is used. fullrate (FR): Fullrate channel or channel mode. frame: A time interval equal to 20 ms (160 samples at an 8 khz sampling rate). gross bitrate: The bitrate of the channel mode selected (22.8 kbs or 11.4 kbs). halfrate (HR): Halfrate channel or channel mode. inband signalling: Signalling for DTX, Link Control, Channel and codec mode modification, etc. carried within the traffic channel. integer lags: A set of lag values having whole sample resolution. interpolating filter: An FIR filter used to produce an estimate of subsample resolution samples, given an input sampled with integer sample resolution. inverse filter: This filter removes the short term correlation from the speech signal. The filter models an inverse frequency response of the vocal tract. lag: The long term filter delay. This is typically the true pitch period, or its multiple or submultiple. Line Spectral Frequencies: (see Line Spectral Pair) Line Spectral Pair: Transformation of LPC parameters. Line Spectral Pairs are obtained by decomposing the inverse filter transfer function A(z) to a set of two transfer functions, one having even symmetry and the other having odd symmetry. The Line Spectral Pairs (also called as Line Spectral Frequencies) are the roots of these polynomials on the z unit circle. LP analysis window: For each frame, the short term filter coefficients are computed using the high pass filtered speech samples within the analysis window. In the adaptive multirate codec, the length of the analysis window is always 240 samples. For each frame, two asymmetric windows SIST EN are301 used704 to generate V7.1.1:2003 two sets of LP coefficient in the 12.2 kbit/s mode. For the other modes, only a single asymmetric window is used to generate a single set of LP coefficients. In the 12.2 kbit/s mode, no samples of the future frames are used (no lookahead). The other modes use a 5 ms lookahead. LP coefficients: Linear Prediction (LP) coefficients (also referred as Linear Predictive Coding (LPC) coefficients) is a generic descriptive term for the short term filter coefficients. mode: When used alone, refers to the source codec mode, i.e., to one of the source codecs employed in the AMR codec. (See also codec mode and channel mode.) openloop pitch search: A process of estimating the near optimal lag directly from the weighted speech input. This is done to simplify the pitch analysis and confine the closedloop pitch search to a small number of lags around the openloop estimated lags. In the adaptive multirate codec, an openloop pitch search is performed in every other subframe. outofband signalling: Signalling on the GSM control channels to support link control. residual: The output signal resulting from an inverse filtering operation. short term synthesis filter: This filter introduces, into the excitation signal, short term correlation which models the impulse response of the vocal tract. perceptual weighting filter: This filter is employed in the analysisbysynthesis search of the codebooks. The filter exploits the noise masking properties of the formants (vocal tract resonances) by weighting the error less in regions near the formant frequencies and more in regions away from them. subframe: A time interval equal to 5 ms (40 samples at 8 khz sampling rate). vector quantization: A method of grouping several parameters into a vector and quantizing them simultaneously. zero input response: The output of a filter due to past inputs, i.e. due to the present state of the filter, given that an input of zeros is applied.
11 9 zero state response:the output of a filter due to the present input, given that no past inputs have been applied, i.e., given that the state information in the filter is all zeroes. 3.2 Symbols For the purposes of the present document, the following symbols apply: ( ) () Az A z ( ) Hz a i = 1 Az ( ) The inverse filter with unquantized coefficients The inverse filter with quantized coefficients The speech synthesis filter with quantized coefficients The unquantized linear prediction parameters (direct form coefficients) The quantified linear prediction parameters The order of the LP model a i m 1 The longterm synthesis filter Bz () Wz () The perceptual weighting filter (unquantized coefficients) γ 1, γ 2 The perceptual weighting factors F () E z Adaptive prefilter T β Az ( / γ n ) Hf () z = The formant postfilter Az ( / γ d ) γ n Control coefficient for the SIST amount EN 301 of the 704 formant V7.1.1:2003 postfiltering γ d Control coefficient for the amount of the formant postfiltering H ( z) Tilt compensation filter γ t t The integer pitch lag nearest to the closedloop fractional pitch lag of the subframe The adaptive prefilter coefficient (the quantified pitch gain) Control coefficient for the amount of the tilt compensation filtering µ = γ t k 1 ' A tilt factor, with k 1 ' being the first reflection coefficient hf The truncated impulse response of the formant postfilter L h The length of hf rh( i) The autocorrelations of hf A zγ ( n ) 1 Az ( γ d ) r The residual signal of the inverse filter A ( zγ ) The inverse filter (numerator) part of the formant postfilter The synthesis filter (denominator) part of the formant postfilter n Impulse response of the tilt compensation filter ht β sc The AGCcontrolled gain scaling factor of the adaptive postfilter α The AGC factor of the adaptive postfilter Hh1( z) Preprocessing highpass filter wi n II LP analysis windows L I 1 ( ) Length of the first part of the LP analysis window wi ( n ) L I 2 ( ) Length of the second part of the LP analysis window wi ( n ) L II 1 ( ) Length of the first part of the LP analysis window wii ( n ) L2 ( II) Length of the second part of the LP analysis window wii ( n )
12 10 rac( k) w () i f 0 f s lag r' ( k) E ac LD k i () i a j () i The autocorrelations of the windowed speech s' Lag window for the autocorrelations (60 Hz bandwidth expansion) The bandwidth expansion in Hz The sampling frequency in Hz The modified (bandwidth expanded) autocorrelations The prediction error in the ith iteration of the Levinson algorithm The ith reflection coefficient The jth direct form coefficient in the ith iteration of the Levinson algorithm F1 ( z) Symmetric LSF polynomial F2 ( z) Antisymmetric LSF polynomial F1 ( z) Polynomial F1 ( z) with root z =1 eliminated F2 ( z) Polynomial F2 ( z) with root z = 1 eliminated q i q ( n q ) i ω i The line spectral pairs (LSPs) in the cosine domain An LSP vector in the cosine domain The quantified LSP vector at the ith subframe of the frame n The line spectral frequencies (LSFs) T ( m x ) A m th order Chebyshev polynomial f1(), i f2() i The coefficients of the polynomials iteh STANDARD F1 ()and z PREVIEW F 2 () z ' ' f1(), i f2() i The coefficients of the polynomials F1 ( z) F2 ( z) f () i The coefficients of either F1 ( z) or F2 ( z) Cx ( ) Sum polynomial of the Chebyshev SIST EN 301 polynomials 704 V7.1.1:2003 x Cosine of angular frequency ω λ k f i [ ] f t f f f Recursion coefficients for the Chebyshev polynomial evaluation The line spectral frequencies (LSFs) in Hz = The vector representation of the LSFs in Hz z ( 2) n The meanremoved LSF vectors at frame n z () ( ) 1 n, ( ) r () 1 ( n ), r ( ) ( ) p r ( 2) ( n 1) f k E LSP 2 n The LSF prediction residual vectors at frame n The predicted LSF vector at frame n The quantified second residual vector at the past frame The quantified LSF vector at quantization index k The LSP quantization error w, i i = 1,, 10, LSPquantization weighting factors d i ( ) hn O k Ot i, i=1,, 3 ( M t ) The distance between the line spectral frequencies f i+1 and f i1 The impulse response of the weighted synthesis filter The correlation maximum of openloop pitch analysis at delay k The correlation maxima at delays ti, i = 1,, 3 i, i, i=1,, 3 The normalized correlation maxima M i and the corresponding delays ti, i = 1,, 3 Az ( / γ 1) HzWz () () = The weighted synthesis filter AzAz () ( / γ ) 2
13 11 Azγ ( 1 ) 1 Azγ ( 2 ) T 1 s' sw s s sf xn ( ) x reslp ( n ) cn ( ) vn ( ) 2, x 2 t yn ( )= vn ( ) hn ( ) yk un ( ) ( ) u n u' T op t min t max The numerator of the perceptual weighting filter The denominator of the perceptual weighting filter The integer nearest to the fractional pitch lag of the previous (1st or 3rd) subframe The windowed speech signal The weighted speech signal Reconstructed speech signal The gainscaled postfiltered signal Postfiltered speech signal (before scaling) The target signal for adaptive codebook search The target signal for algebraic codebook search The LP residual signal The fixed codebook vector The adaptive codebook vector The filtered adaptive codebook vector The past filtered excitation The excitation signal The emphasized adaptive codebook vector The gainscaled emphasized excitation signal The best openloop lag Minimum lag search value Maximum lag search value Correlation term to be maximized in the adaptive codebook search Rk ( ) b 24 The FIR filter for interpolating the normalized correlation term Rk ( ) Rk ( ) t The interpolated value of Rk ( ) for the integer delay k and fraction t b 60 The FIR filter for interpolating the past excitation signal un ( ) to yield the adaptive codebook A k vector vn ( ) Correlation term to be maximized in the algebraic codebook search at index k C k The correlation in the numerator of A k at index k E D k t The energy in the denominator of A k at index k d = H x2 The correlation between the target signal x filtered target 2 and the impulse response ( ) H The lower triangular Toepliz convolution matrix with diagonal ( ) h() 1,, h( 39) Φ=HH The matrix of correlations of hn ( ) dn ( ) The elements of the vector d φ( i, j) c k C m i ϑ i The elements of the symmetric matrix Φ The innovation vector The correlation in the numerator of A k The position of the ith pulse The amplitude of the ith pulse hn, i.e., backward h 0 and lower diagonals
14 12 N p The number of pulses in the fixed codebook excitation E D The energy in the denominator of A k resltp bn ( ) sb d Sign extended backward filtered target φ ' (, ij) The normalized longterm prediction residual The signal used for presetting the signs in algebraic codebook search The sign signal for the algebraic codebook search The modified elements of the matrix Φ, including sign information z t, zn ( ) The fixed codebook vector convolved with hn ( ) En ( ) E ~ En ( ) [ b b b b ] The meanremoved innovation energy (in db) The mean of the innovation energy The predicted energy The MA prediction coefficients R ( k) E I The quantified prediction error at subframe k The mean innovation energy Rn ( ) The prediction error of the fixedcodebook gain quantization E Q The quantization error of the fixedcodebook gain quantization en ( ) The states of the synthesis filter 1 Az ( ) ew η g c g c g c g p g p γ gc g c g c The perceptually weighted error of the analysisbysynthesis search The gain scaling factor for the emphasized excitation The fixedcodebook gain The predicted fixedcodebook SIST EN gain V7.1.1:2003 The quantified fixed codebook gain The adaptive codebook gain The quantified adaptive codebook gain = A correction factor between the gain g c and the estimated one g c γ gc γ sc The optimum value for γ gc Gain scaling factor 3.3 Abbreviations For the purposes of the present document, the following abbreviations apply. Further GSM related abbreviations may be found in GSM [1]. ACELP AGC AMR CELP EFR FIR FR HR ISPP LP LPC LSF LSP Algebraic Code Excited Linear Prediction Adaptive Gain Control Adaptive MultiRate Code Excited Linear Prediction Enhanced Full Rate Finite Impulse Response Full Rate Half Rate Interleaved SinglePulse Permutation Linear Prediction Linear Predictive Coding Line Spectral Frequency Line Spectral Pair
15 13 LTP MA Long Term Predictor (or Long Term Prediction) Moving Average 4 Outline description The present document is structured as follows: Section 4.1 contains a functional description of the audio parts including the A/D and D/A functions. Section 4.2 describes the conversion between 13bit uniform and 8bit Alaw or µ law samples. Sections 4.3 and 4.4 present a simplified description of the principles of the AMR codec encoding and decoding process respectively. In subclause 4.5, the sequence and subjective importance of encoded parameters are given. Section 5 presents the functional description of the AMR codec encoding, whereas clause 6 describes the decoding procedures. In section 7, the detailed bit allocation of the AMR codec is tabulated. 4.1 Functional description of audio parts The analoguetodigital and digitaltoanalogue conversion will in principle comprise the following elements: 1) Analogue to uniform digital PCM microphone; input level adjustment device; input antialiasing filter; samplehold device sampling at 8 khz; analoguetouniform digital conversion SIST EN to 13bit representation. V7.1.1:2003 The uniform format shall be represented in two's complement. 2) Uniform digital PCM to analogue conversion from 13bit/8 khz uniform PCM to analogue; a hold device; reconstruction filter including x/sin( x ) correction; output level adjustment device; earphone or loudspeaker. In the terminal equipment, the A/D function may be achieved either by direct conversion to 13bit uniform PCM format; or by conversion to 8bit Alaw or µ law compounded format, based on a standard Alaw or µ law codec/filter according to ITUT Recommendations G.711 [8] and G.714, followed by the 8bit to 13bit conversion as specified in subclause For the D/A operation, the inverse operations take place. In the latter case it should be noted that the specifications in ITUT G.714 (superseded by G.712) are concerned with PCM equipment located in the central parts of the network. When used in the terminal equipment, the present document does not on its own ensure sufficient outofband attenuation. The specification of outofband signals is defined in GSM [2] in clause 2.