G191 E - Hackipedia.org
Contents
1. L_msu0 L_v3 vl v2 L_msuNs L_v3 vl v2 msu_r L_v3 vl v2 s_and vl1 v2 s or vl v2 s_xor v1 v2 lshl v1 v2 Multiplies v1 by v2 and shifts the result left by 1 Adds the 32 bit result to L_v3 with saturation Rounds the 16 least significant bits of the result into the 16 most significant bits with saturation and shifts the result right by 16 Returns a 16 bit result mac_r L v3 vl v2 round L mac L v3 vl v2 extract_h L_add L_add L_ v3 L_mult vl v2 32768 Multiplies v1 by v2 and shifts the result left by 1 Subtracts the 32 bit result from L_v3 with saturation returns a 32 bit result L msu L v3 vl v2 L_sub L_v3 L mult v1 v2 Multiplies vl by v2 without left shift Subtracts the 32 bit result from L_v3 with saturation returning a 32 bit result L msu L v3 vl v2 L sub L v3 L mult0 vil v2 Multiplies v1 by v2 and shifts the result left by 1 Subtracts the 32 bit result from L_v3 without saturation returns a 32 bit result Generates carry and overflow values L msuNs L v3 vl v2 L sub c L v3 L mult vl v2 Multiplies v1 by v2 and shifts the result left by 1 Subtracts the 32 bit result from L v3 with saturation Rounds the 16 least significant bits of the result into the 16 bits with saturation and shifts the result right by 16 Returns a 16 bit result msu_r L v3 vl v2 round L msu L v3 vl v2 extract_h L_add L_s2. B 10 Because the Module is licensed free of charge there is no warranty for the Module to the extent permitted by applicable law Except when otherwise stated in writing the copyright holders and or other parties provide the Module as is without warranty of any kind either expressed or implied including but not limited to the implied warranties of merchantability and fitness for a particular purpose The entire risk as to the quality and performance of the Module is with you Should the Module prove defective you assume the cost of all necessary servicing repair or correction B 11 In no event unless required by applicable law or agreed to in writing will any copyright holder or any other party who may modify and or redistribute the Module as permitted above be liable to you for damages including any general special incidental or consequential damages arising out of the use or inability to use the Module including but not limited to loss of data or data being rendered inaccurate or losses sustained by you or third parties or a failure of the Module to operate with any other modules even if such holder or other party has been advised of the possibility of such damages 20 Rec ITU T G 191 03 2010 Series A Series D Series E Series F Series G Series H Series I Series J Series K Series L Series M Series N Series O Series P Series Q Series R Series S Series T Series U Series V Series X Series Y Series Z S
3. f Appendix I G 711 Packet Loss Concealment module Name lowcfe c Associated header file lowcfe h Functions included g71lplc_construct LowcFE Constructor g711plc_dofe generate the synthetic signal g711plc_addtohistory a good frame was received and decoded add the frame to history buffer g G 726 module Name g726 c Associated header file g726 h Functions included G726_encode ITU T G 726 encoder at 40 32 24 and 16 kbit s G726_decode ITU T G 726 decoder at 40 32 24 and 16 kbit s h Modulated noise reference unit module Name mnru c Associated header file mnru h Functions included MNRU_process module for addition of modulated noise to a vector of samples according to Rec ITU T P 810 for both the narrow band and the wideband models i Speech voltmeter module Name sv p56 c Associated header file sv p56 h Functions included init_speech_voltmeter initializes a speech voltmeter state variable speech_voltmeter measurement of the active speech level of data in a buffer according to Rec ITU T P 56 6 Rec ITU T G 191 03 2010 j Module with UGST utilities Name ugst utl c Associated header file ugst utl h Functions included scale sh2f1l 16bi sh2f1 15bi sh2fl 14bi sh2f1_ 13bi sh2f1l 12bi sh2f1 sh2fl_alt fl2sh_16bi fl2sh_15bi fl2sh_14bi fl2sh_13bi fl2sh_12bi fl2sh serialize_left_justified serialize_right_justified parallelize_left_justified parallelize_right_justi
4. International Telecommunication Union ITU T G 191 TELECOMMUNICATION 03 2010 STANDARDIZATION SECTOR OF ITU SERIES G TRANSMISSION SYSTEMS AND MEDIA DIGITAL SYSTEMS AND NETWORKS International telephone connections and circuits Software tools for transmission systems Software tools for speech and audio coding standardization Recommendation ITU T G 191 ITU T G SERIES RECOMMENDATIONS TRANSMISSION SYSTEMS AND MEDIA DIGITAL SYSTEMS AND NETWORKS INTERNATIONAL TELEPHONE CONNECTIONS AND CIRCUITS Transmission planning and the E model General Recommendations on the transmission quality for an entire international telephone connection General characteristics of national systems forming part of international connections General characteristics of the 4 wire chain formed by the international circuits and national extension circuits General characteristics of the 4 wire chain of international circuits international transit General characteristics of international telephone circuits and national extension circuits Apparatus associated with long distance telephone circuits Transmission plan aspects of special circuits and connections using the international telephone connection network Protection and restoration of transmission systems Software tools for transmission systems GENERAL CHARACTERISTICS COMMON TO ALL ANALOGUE CARRIER TRANSMISSION SYSTEMS INDIVIDUAL CHARACTERISTICS OF INTERNATIONAL CARRIER TELEPHONE SYS
5. which meets every four years establishes the topics for study by the ITU T study groups which in turn produce Recommendations on these topics The approval of ITU T Recommendations is covered by the procedure laid down in WTSA Resolution 1 In some areas of information technology which fall within ITU T s purview the necessary standards are prepared on a collaborative basis with ISO and IEC NOTE In this Recommendation the expression Administration is used for conciseness to indicate both a telecommunication administration and a recognized operating agency Compliance with this Recommendation is voluntary However the Recommendation may contain certain mandatory provisions to ensure e g interoperability or applicability and compliance with the Recommendation is achieved when all of these mandatory provisions are met The words shall or some other obligatory language such as must and the negative equivalents are used to express requirements The use of such words does not suggest that compliance with the Recommendation is required of any party INTELLECTUAL PROPERTY RIGHTS ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right ITU takes no position concerning the evidence validity or applicability of claimed Intellectual Property Rights whether asserted by ITU members or others outside of the Recommendation development pro
6. FIR filters available in the rate change module mnrudemo c on the use of the narrow band and wideband modulated noise reference unity ITU T P 81 module 2 Rec ITU T G 191 03 2010 spdemo c g71lliplc c reverb c truncate c freqresp c stereoop c on the use of the serialization and parallelization routines of the utility module on the use of Appendix I G 711 Packet Loss Concealment module on the use of the reverberation module on the use of the bitstream truncation module on the use of the frequency response computation tool on the use of stereo file operations NOTE The module for the Basic Operators does not have a demo program but it is supplemented by two tools one to evaluate program ROM complexity for fixed point code basop_cnt c and another to evaluate complexity including program ROM of floating point implementations lc_example c Both reside in the Basic Operators module b Rate change module with FIR finite impulse response routines Name firflt c Associated header file firflt h Functions included delta_sm_16khz_ init hq down 2 to 1 init hq down 3 to 1 init hq up_1 to 2 init hq _up_1 to 3 init irs 8khz init irs 16khz init linear phase pb 2 to 1 init linear phase pb 1 to 2 init linear phase pb 1 to 1 init mod_irs 16khz init mod_irs 48khz init psophometric 8khz init p341_ 16khz init rx mod irs 16khz init rx mod irs 8khz init
7. ITU T Software Tool Library STL are free software they can be redistributed and or modified under the terms of the ITU T software tools General Public License of Annex B as published by the ITU T this applies to any of the versions of the modules in the STL The STL has been carefully tested and it is believed that both the modules and the example programs on their usage conform to their description documents Nevertheless the ITU T STL is provided as is in the hope that it will be useful but without any warranty The STL is intended to help the scientific community to achieve new standards in telecommunications more efficiently and for such must not be sold entirely or in parts The original developers except where otherwise noted retain ownership of their copyright and allow their use under the terms and conditions of the ITU T software tools General Public License Rec ITU T G 191 03 2010 1 Annex A List of software tools available This annex forms an integral part of this Recommendation This annex contains a list with a short description of the software tools available in the ITU T Software Tool Library This is referred to in the associated documentation as the Software Tool Library release 2009 or STL2009 All the routines in the modules are written in C a Example programs available Associated header file ugstdemo h The following programs are examples of the use of the modules g711demo c on the use
8. L40 vl norm_L40 L40 v1 Multiplies the 2 signed 16 bit variables vl and v2 without saturation control on 40 bits Any detected overflow on 40 bits will exit execution The operation is performed in fractional mode vl and v2 are supposed to be in 1Q15 format The result is produced in 9Q31 format Equivalent to L40 add L40 vl L40 mult v2 v3 Equivalent to L40 sub L40 vl L40 mult v2 v3 Logically shifts left the 40 bit variable L40 vl by v2 positions if v2 is negative L40 v1 is shifted to the least significant bits by v2 positions with insertion of 0 at the most significant bit if v2 is positive L40_v1 is shifted to the most significant bits by v2 positions without saturation control Logically shifts right the 40 bit variable L40 vl by v2 positions if v2 is positive L40_ v1 is shifted to the least significant bits by v2 positions with insertion of 0 at the most significant bit if v2 is negative L40_ vl is shifted to the most significant bits by v2 positions without saturation control Returns the bits 31 16 of L40_ v1 Returns the bits 15 00 of L40_v1 Equivalent to extract_h L_saturate40 L40_ round L40 v1 Returns the bits 31 00 of L40_v1 Rec ITU T G 191 03 2010 15 L_saturate40 L40_v1 L40_deposit_h v1 L40_deposit_1 v1 L40_deposit32 L_v1 L40_round L40_ v1 mac_r40 L40 vl v2 v3 msu_r40 L40 vl v2 v3 Mpy 32 16 ss L vl v2 L v3
9. T G 722 frame erasure concealments G 722 PLCO and PLC1 program ROM counting floating point complexity evaluation stereo operation and bitstream truncation There have been also updates such as addition of new filters to rate change tools more options added for frequency response calculations adaptation of error insertion device to embedded variable bitstream addition of impulse responses to the reverberation processing tool and update to make ITU T G 722 ITU T G 192 compliant Corrections of several known problems have also been implemented The associated STL user s manual has also been updated This Recommendation includes an electronic attachment containing STL2009 Software Tool Library and manual History Edition Recommendation Approval Study Group 10 ITU T G 191 1993 03 12 XV 2 0 ITU T G 191 1996 11 11 15 3 0 ITU T G 191 2000 11 17 16 4 0 ITU T G 191 2005 09 13 16 5 0 ITU T G 191 2010 03 29 16 Rec ITU T G 191 03 2010 i FOREWORD The International Telecommunication Union ITU is the United Nations specialized agency in the field of telecommunications information and communication technologies ICTs The ITU Telecommunication Standardization Sector ITU T is a permanent organ of ITU ITU T is responsible for studying technical operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis The World Telecommunication Standardization Assembly WTSA
10. form IIR filtering routine Initialize a direct form IIR filter structure for a 1 1 DC removal filtering clear state variables needed only if another signal should be processed with the same filter for a direct form structure deallocate filter memory for a direct form state variable structure initializes the error pattern generator for single bit errors burst bit errors or single frame erasures initializes the burst frame erasure pattern generator reinitializes the burst frame erasure pattern generator generates a bit error sequence with properties defined by open_eid generates a random frame erasure sequence with properties defined by open eid y generates a burst frame erasure sequence with properties defined by open_burst_eid modifies the input data bits according to the error pattern stored in a buffer frame erasure module frees memory allocated to the EID state variable buffer Rec ITU T G 191 03 2010 5 alaw_compress compands 1 vector of linear PCM samples to A law uses 13 Most Significant Bits MSBs from input and 8 Least Significant Bits LSBs on output alaw_expand expands 1 vector of A law samples to linear PCM uses 8 LSBs from input and 13 MSBs on output ulaw_compress compands 1 vector of linear PCM samples to p law uses 14 MSBs from input and 8 LSBs on output ulaw_expand expands 1 vector of law samples to linear PCM uses 8 LSBs from input and 14 MSBs on output
11. this General Public License except that you may choose to grant warranty protection to some or all third parties at your option e if the modified module normally reads commands interactively when run you must cause it when started running for such interactive use in the simplest and most usual way to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty or else saying that you provide a warranty and that users may redistribute the module under these conditions and telling the user how to view a copy of this General Public License You may charge a fee for the physical act of transferring a copy and you may at your option offer warranty protection in exchange for a fee Mere aggregation of another independent work with the Module or its derivative on a volume of a storage or distribution medium does not bring the other work under the scope of these terms B 4 You may copy and distribute the Module or a portion or derivative of it under B 2 in object code or executable form under the terms of B 1 and B 2 provided that you also do one of the following e accompany it with the complete corresponding machine readable source code which must be distributed under the terms of B 1 and B 2 or accompany it with a written offer valid for at least three years to give any third party free except for a nominal charge for the cost of distribution a complete machine readable c
12. tia_irs 8khz init ht_irs 16khz_ init msin_16khz init bp5k_16khz_ init initialize 16 kHz 1 1 ASM weighting filter initialize 2 1 low pass down sampling filter initialize 3 1 low pass down sampling filter initialize 1 2 low pass up sampling filter initialize 1 3 low pass up sampling filter initialize 8 kHz P 48 IRS weighting filter initialize 16 kHz P 48 IRS weighting filter initialize 2 1 bandpass down sampling filter initialize 1 2 bandpass up sampling filter initialize 1 1 bandpass filter initialize 16 kHz send side modified IRS weighting filter initialize 48 kHz send side modified IRS weighting filter initialize 1 1 ITU T 0 41 psophometric weighting filter initialize 1 1 ITU T P 341 send part weighting filter for data sampled at 16 kHz initialize 16 kHz modified IRS receive side weighting filter initialize 8 kHz modified IRS receive side weighting filter initialize 8 kHz IRS weighting filter using the TIA coefficients initialize 16 kHz IRS weighting filter with a half tilt inclination within the ITU T P 48 mask initialize mobile station weighting filter initialize 50 Hz to 5 kHz bandpass filter 16 kHz sampling Rec ITU T G 191 03 2010 3 bp100_ 5k 16khz init bp14k_ 32khz init bp20k_ 48khz init LP1p5 48kHz init LP35 48kHz init LP7_ 48kHz init LP10 48kHz init LP12 48kHz init LP14_ 48kHz init LP20 48kHz init initialize a 100 Hz to 5 kHz bandpass filter
13. 1 by v2 positions if v2 is negative L_v1 is shifted to the least significant bits by v2 positions with insertion of 0 at the most significant bit if v2 is positive L_v1 is shifted to the most significant bits by v2 positions without saturation control Logically shifts right the 32 bit variable L vl by v2 positions if v2 is positive L_v1 is shifted to the least significant bits by v2 positions with insertion of 0 at the most significant bit if v2 is negative L_v1 is shifted to the most significant bits by v2 positions without saturation control Returns the 16 most significant bits of L_v1 Returns the 16 least significant bits of L_v1 Rounds the lower 16 bits of the 32 bit input number into the most significant 16 bits with saturation Shifts the resulting bits right by 16 and returns the 16 bit number round L_vl extract_h L_add L_ v1 32768 Deposits the 16 bit v1 into the 16 bit most significant bits of the 32 bit output The 16 least significant bits of the output are zeroed Deposits the 16 bit v1 into the 16 bit least significant bits of the 32 bit output The 16 most significant bits of the output are sign extended Performs the 32 bit addition with carry No saturation Generates carry and overflow values The carry and overflow values are binary variables which can be tested and assigned values Performs the 32 bit subtraction with carry borrow Generates carry borrow and overflow v
14. 16 kHz sampling initialize a 50 Hz to 14 kHz bandpass filter 32 kHz sampling initialize a 20 Hz to 20 kHz bandpass filter 48 kHz sampling initialize a low pass filter with a cut off frequency of 1 5 kHz 48 kHz sampling initialize a low pass filter with a cut off frequency of 3 5 kHz 48 kHz sampling initialize a low pass filter with a cut off frequency of 7 kHz 48 kHz sampling initialize a low pass filter with a cut off frequency of 10 kHz 48 kHz sampling initialize a low pass filter with a cut off frequency of 12 kHz at 48 kHz sampling initialize a low pass filter with a cut off frequency of 14 kHz at 48 kHz sampling initialize a low pass filter with a cut off frequency of 20 kHz 48 kHz sampling hq_kernel FIR filtering function hq_reset clear state variables hq_free deallocate FIR filter memory c Rate change module with IIR routines Name iirflt c Associated header file iirflt h Functions included stdpcm_kernel stdpcm_16khz init stdpcm_1 to 2 init stdpcm_2 to 1 init n tdpcm_reset stdpcm_ free cascade_iir kernel iir_G712_8khz init 4 Rec ITU T G 191 03 2010 parallel form IIR kernel filtering routine initialization of a parallel form IIR standard PCM filter for input and output data at 16 kHz as stdpcm_16khz_init but needs input with sampling frequency of 8 kHz and returns data at 16 kHz as stdpcm_16khz_init but needs input with sampling f
15. 2 s_min v1 v2 norm_s v1 L add L v1 L v2 L sub L v1 L v2 L_abs L_v1 L shl L_v1 v2 L_shr L_v1 v2 L negate L_v1 L max L vl L v2 Arithmetically shifts the 16 bit input v1 left by v2 positions Zero fills the v2 LSB of the result If v2 is negative arithmetically shifts v1 right by v2 with sign extension Saturates the result in case of underflows or overflows Arithmetically shifts the 16 bit input v1 right v2 positions with sign extension If v2 is negative arithmetically shifts v1 left by v2 and zero fills the v2 LSB of the result shr v1 v2 shl vl1 v2 Saturates the result in case of underflows or overflows Negates v1 with saturation saturate in the case when input is 32768 negate v1l sub 0 v1 Compares two 16 bit variables vl and v2 and returns the maximum value Compares two 16 bit variables vl and v2 and returns the minimum value Produces the number of left shifts needed to normalize the 16 bit variable v1 for positive values on the interval with minimum of 16384 and maximum 32767 and for negative values on the interval with minimum of 32768 and maximum of 16384 in order to normalize the result the following operation must be done norm_vl shl vl norm_s v1 32 bit addition of the two 32 bit variables L_v1 L_v2 with overflow control and saturation the result is set at 2147483647 when overflow occurs or at 2147483648 when underf
16. ERIES OF ITU T RECOMMENDATIONS Organization of the work of ITU T General tariff principles Overall network operation telephone service service operation and human factors Non telephone telecommunication services Transmission systems and media digital systems and networks Audiovisual and multimedia systems Integrated services digital network Cable networks and transmission of television sound programme and other multimedia signals Protection against interference Construction installation and protection of cables and other elements of outside plant Telecommunication management including TMN and network maintenance Maintenance international sound programme and television transmission circuits Specifications of measuring equipment Terminals and subjective and objective assessment methods Switching and signalling Telegraph transmission Telegraph services terminal equipment Terminals for telematic services Telegraph switching Data communication over the telephone network Data networks open system communications and security Global information infrastructure Internet protocol aspects and next generation networks Languages and general software aspects for telecommunication systems Printed in Switzerland Geneva 2010
17. TEMS ON METALLIC LINES GENERAL CHARACTERISTICS OF INTERNATIONAL CARRIER TELEPHONE SYSTEMS ON RADIO RELAY OR SATELLITE LINKS AND INTERCONNECTION WITH METALLIC LINES COORDINATION OF RADIOTELEPHONY AND LINE TELEPHONY TRANSMISSION MEDIA AND OPTICAL SYSTEMS CHARACTERISTICS DIGITAL TERMINAL EQUIPMENTS DIGITAL NETWORKS DIGITAL SECTIONS AND DIGITAL LINE SYSTEM MULTIMEDIA QUALITY OF SERVICE AND PERFORMANCE GENERIC AND USER RELATED ASPECTS TRANSMISSION MEDIA CHARACTERISTICS DATA OVER TRANSPORT GENERIC ASPECTS PACKET OVER TRANSPORT ASPECTS ACCESS NETWORKS For further details please refer to the list of ITU T Recommendations G 100 G 199 G 100 G 109 G 110 G 119 G 120 G 129 G 130 G 139 G 140 G 149 G 150 G 159 G 160 G 169 G 170 G 179 G 180 G 189 G 190 G 199 G 200 G 299 G 300 G 399 G 400 G 449 G 450 G 499 G 600 G 699 G 700 G 799 G 800 G 899 G 900 G 999 G 1000 G 1999 G 6000 G 6999 G 7000 G 7999 G 8000 G 8999 G 9000 G 9999 Recommendation ITU T G 191 Software tools for speech and audio coding standardization Summary This revision only introduces changes to ITU T G 191 Annex A which describes the ITU T Software Tools STL containing a high quality portable C code library for speech processing applications This new release of the STL dubbed STL2009 incorporates many changes in software and manual The new sets of tools added are ITU T G 728 fixed floating point executables basic ITU
18. _h v3_1 L40 _ shr r L40 v1 v2 L40_shl_r L40 vl v2 16 Rec ITU T G 191 03 2010 If L40_v1 is greater than 2147483647 returns 2147483647 If L40_v1 is lower than 2147483648 returns 2147483648 If not equivalent to L Extract40 L40_ v1 Deposits the 16 bit variable v1 in the bits 31 16 of the return value the return value bits 15 0 are set to 0 and the bits 39 32 sign extend v1 sign bit Deposits the 16 bit variable v1 in the bits 15 0 of the return value the return value bits 39 16 sign extend v1 sign bit Deposits the 32 bit variable L_v1 in the bits 31 0 of the return value the return value bits 39 32 sign extend L_vl sign bit Performs a rounding to the infinite on the 40 bit variable L40 vl 32768 is added to L40 v1 without saturation control on 40 bits Any detected overflow on 40 bits will exit execution The end result 16 LSB are cleared to 0 Equivalent to round40 L40 mac L40 vl v2 v3 Equivalent to round40 L40 msu L40 vl v2 v3 Multiplies the 2 signed values L_v1 32 bit and v2 16 bit with saturation control on 48 bits The operation is performed in fractional mode When L v1 is in 1Q31 format and v2 is in 1Q15 format the result is produced in 1Q47 format L_v3_h bears the 32 most significant bits while v3_1 bears the 16 least significant bits Arithmetically shifts the 40 bit variable L40 vl by v2 positions to the least significan
19. alues No saturation The carry and overflow values are binary variables which can be tested and assigned values shr_r vl v2 Same as shr vl v2 but with rounding Saturates the result in case of underflows or overflows if v2 is strictly greater than zero then if sub sht shr vil v2 1 shr vl sub v2 1 0 then shr_r v1 v2 shr vl v2 else shr_r vl v2 add shr v1 v2 1 On the other hand if v2 is lower than or equal to zero then shr_r vl v2 shr vl v2 shl_r vl v2 Same as shl vl v2 but with rounding Saturates the result in case of underflows or overflows shl_r vl v2 shr_r vl v2 In the previous version of the STL basic operators this operator is called shift_r v1 v2 both names can be used L_shr_r L_vl v2 Same as L_ shr vl v2 but with rounding Saturates the result in case of underflows or overflows if v2 is strictly greater than zero then if L sub L_shl L_shr Lvl v2 1 L_shr L v1 sub v2 1 then L shr r Lvl v2 L_shr Lvl v2 else L_shr r Lvl v2 L add L shr L v1 v2 1 On the other hand if v2 is less than or equal to zero then L shr_r Lvl v2 L_shr L v1 v2 L shl_r L_v1 v2 Same as L_shl L_v1 v2 but with rounding Saturates the result in case of underflows or overflows L shift_r Lvl v2 L_shr_r Lvl v2 In the previous version of the STL basic operators this operator is called L_shift_r L_vl v2 both name
20. ations may not perfectly conform to related ITU T Recommendations which may delay ITU T standardization processes c ITU T Recommendations may leave scope for different implementations d New speech and audio coding standards are increasing in complexity leading to non bitexact specifications furthermore appropriate testing procedures to assure interoperability of different implementations are needed The need for a common set of tools has been recognized in past ITU T standardization activities of speech algorithms As a consequence a library of portable interworkable and reliable software routines has been established 2 Software tools To clarify the use of the set of software tools arranged as a software tool library the ITU T makes the following recommendations 1 The software tools specified in Annex A should be used as building modules of signal processing blocks to be used in the process of generation of ITU T Recommendations particularly those concerned with speech and audio coding algorithms 2 Some of the tools shall be used in procedures for the verification of interoperability of ITU T standards mainly of speech and audio coding algorithms whose description is in terms of non bitexact specifications 3 The use of these modules should be made strictly in accordance with the technical instructions of their attached documentation and should respect the following terms 3 License and copyright The modules in the
21. cess As of the date of approval of this Recommendation ITU had not received notice of intellectual property protected by patents which may be required to implement this Recommendation However implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at http www itu int ITU T ipr ITU 2010 All rights reserved No part of this publication may be reproduced by any means whatsoever without the prior written permission of ITU ii Rec ITU T G 191 03 2010 CONTENTS Page 1 RTT ell oc a et e gk Sta ial ganda aan aaa at ca 1 2 DOLL W Ale OO IG a c cops ccatGoatehcotun ar a tence a A T a e 1 3 LACENSE AaNd COPY MON 2 hive ise Aas deeded wikcaa aa Sane adons aia aad As ees 1 Annex A List of software tools available eee eeesscescesseceesceceecceseessesseeceesenceeaceseeseeenees 2 Annex B ITU T software tools General Public License cecceeceecceeseeeteceseeeeecenseeenseenes 19 Electronic attachment STL2009 Software Tool Library and manual Rec ITU T G 191 03 2010 iii Recommendation ITU T G 191 Software tools for speech and audio coding standardization 1 General In the process of generating speech and audio coding standards the following situations often happen a In many cases experimental results generated with different software tools may not be directly compared b Software tools used by different organiz
22. elete m G 727 module Name g727 c Associated header file g727 h Functions included G727_encode G727_decode n Basic Operators ITU T G 722 wideband speech encoder at 64 kbit s ITU T G 722 wideband speech decoder at 64 56 and 48 kbit s initialization of the ITU T G 722 encoder state variable initialization of the ITU T G 722 decoder state variable GSM 06 10 full rate RPE LTP speech encoder at 13 kbit s GSM 06 10 full rate RPE LTP speech decoder at 13 kbit s initialize memory for the RPE LTP state variables release memory previously allocated for the RPE LTP state variables ITU T G 727 encoder at 40 32 24 and 16 kbit s ITU T G 727 decoder at 40 32 24 and 16 kbit s Name basop32 c enh1632 c enh40 c Associated header file stl h Variable definitions vl v2 16 bit variables e L vl L v2 L_v3 32 bit variables e L40 v1 L40 v2 L40 _ v3 40 bit variables Functions included add v1 v2 sub v1 v2 abs_s v1 8 Rec ITU T G 191 03 2010 Performs the addition vl1 v2 with overflow control and saturation the 16 bit result is set at 32767 when overflow occurs or at 32768 when underflow occurs Performs the subtraction vl v2 with overflow control and saturation the 16 bit result is set at 32767 when overflow occurs or at 32768 when underflow occurs Absolute value of v1 If vl is 32768 returns 32767 shl v1 v2 shr v1 v2 negate v1 s_max v1 v
23. ense Agreement applies to any module or other work related to the ITU T Software Tool Library and developed by the User s Group on Software Tools The term Module below refers to any such module or work and a work based on the Module means either the Module or any work containing the Module or a portion of it either verbatim or with modifications Each licensee is addressed as you B 2 You may copy and distribute verbatim copies of the Module s source code as you receive it in any medium provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty keep intact all the notices that refer to this General Public License and to the absence of any warranty and give any other recipients of the Module a copy of this General Public License along with the Module You may charge a fee for the physical act of transferring a copy B 3 You may modify your copy or copies of the Module or any portion of it and copy and distribute such modifications under the terms of B 1 provided that you also do the following e cause the modified files to carry prominent notices stating that you changed the files and the date of any change and e cause the whole of any work that you distribute or publish that in whole or in part contains the Module or any part thereof either with or without modifications to be licensed at no charge to all third parties under the terms of
24. ense to do so and all its terms and conditions B 7 Each time you redistribute the Module or any work based on the Module the recipient automatically receives a license from the original licensor to copy distribute or modify the Module subject to these terms and conditions You may not impose any further restrictions on the recipients exercise of the rights granted herein B 8 The ITU T may publish revised and or new versions of this General Public License from time to time Such new versions will be similar in spirit to the present version but may differ in detail to address new problems or concerns Each version is given a distinguishing version number If the Module specifies a version number of the license which applies to it and any later version you have the option of following the terms and conditions either of that version or of any later version published by the ITU T If the Module does not specify a version number of the license you may choose any version ever published by the ITU T B 9 If you wish to incorporate parts of the Module into other free modules whose distribution conditions are different write to the author to ask for permission For software which is copyrighted by the ITU T write to the ITU T Secretariat exceptions may be made for this This decision will be guided by the two goals of preserving the free status of all derivatives of this free software and of promoting the sharing and reuse of software generally
25. fied gain loss insertion algorithm conversion of two s complement 16 bit integer to floating point conversion of two s complement 15 bit integer to floating point conversion of two s complement 14 bit integer to floating point conversion of two s complement 13 bit integer to floating point conversion of two s complement 12 bit integer to floating point generic function for conversion from integer to floating point alternate faster implementation of sh2fl with compulsory range conversion conversion of floating point data to two s complement 16 bit integer conversion of floating point data to two s complement 15 bit integer conversion of floating point data to two s complement 14 bit integer conversion of floating point data to two s complement 13 bit integer conversion of floating point data to two s complement 12 bit integer generic function for conversion from floating point to integer serialization for left justified data serialization for right justified data parallelization for left justified data parallelization for right justified data Rec ITU T G 191 03 2010 k G 722 module Name g722 c Associated header file g722 h Functions included G722 encode G722 decode g722 reset_encoder g722 reset_decoder 1 RPE LTP module Name rpeltp c Associated header file rpeltp h Functions included rpeltp_encode rpeltp_decode rpeltp_init rpeltp_d
26. is shifted to the most significant bits by v2 positions without saturation control on 40 bits Any detected overflow on 40 bits will exit execution Arithmetically shifts right the 40 bit variable L40 v1 by v2 positions if v2 is positive L40 _v1 is shifted to the least significant bits by v2 positions with extension of the sign bit if v2 is negative L40 v1 is shifted to the most significant bits by v2 positions without saturation control on 40 bits Any detected overflow on 40 bits will exit execution Negates the 40 bit variable L40 v1 without saturation control on 40 bits Any detected overflow on 40 bits will exit execution L40_max L40 vl L40_v2 L40_min L40 vl L40_v2 norm_L40 L40_ v1 L40_mult v1 v2 L40_mac L40 vl v2 v3 L40 msu L40 vl v2 v3 L40_1shl L40_ vl v2 L40_lshr L40_ v1 v2 Extract40_H L40_v1 Extract40_L L40_v1 round40 L40_v1 L_Extract40 L40_v1 Compares two 40 bit variables L40 v1 and L40 v2 and returns the maximum value Compares two 40 bit variables L40 v1 and L40 v2 and returns the minimum value Produces the number of left shifts needed to normalize the 40 bit variable L40 v1 for positive values on the interval with minimum of 1073741824 and maximum 2147483647 and for negative values on the interval with minimum of 2147483648 and maximum of 1073741824 in order to normalize the result the following operation must be done L40 norm_vl L40_ shl
27. l with minimum of 2147483648 and maximum of 1073741824 in order to normalize the result the following operation must be done L_norm_vl L_shl L_ v1 norm 1 L_v1 L_mult implements the 32 bit result of the multiplication of vl times v2 with one shift left i e L mult v1 v2 L_shl vl v2 1 Note that L mult 32768 32768 2147483647 L_mult0 implements the 32 bit result of the multiplication of vl times v2 without left shift 1 e L mult v1 v2 vl v2 Performs the multiplication of v1 by v2 and gives a 16 bit result which is scaled i e mult v1 v2 extract_1 L_shr vl times v2 15 Note that mult 32768 32768 32767 Same as mult but with rounding i e mult _r vl v2 extract_1 L_shr v1 v2 16384 15 and mult _r 32768 32768 32767 Multiplies v1 by v2 and shifts the result left by 1 Adds the 32 bit result to L_v3 with saturation returns a 32 bit result L mac L v3 vl v2 L add L v3 L mult vl v2 Multiplies v1 by v2 without left shift Adds the 32 bit result to L_v3 with saturation returning a 32 bit result L mac L_ v3 vl v2 L_add vL_v3 L_ multo vv1 v2 Multiplies v1 by v2 and shifts the result left by 1 Adds the 32 bit result to L_v3 without saturation returns a 32 bit result Generates carry and overflow values L_macNs L_v3 vi v2 L add c L v3 L mult v1 v2 mac_r L_v3 vl v2 L msu L_v3 vl v2
28. low occurs 32 bit subtraction of the two 32 bit variables L_v1l L_v2 with overflow control and saturation the result is set at 2147483647 when overflow occurs or at 2147483648 when underflow occurs Absolute value of L vl with L_abs 2147483648 2147483647 Arithmetically shifts the 32 bit input L_v1 left v2 positions Zero fills the v2 LSB of the result If v2 is negative arithmetically shifts L_v1 right by v2 with sign extension Saturates the result in case of underflows or overflows Arithmetically shifts the 32 bit input L_v1 right v2 positions with sign extension If v2 is negative arithmetically shifts L_vl left by v2 and zero fills the v2 LSB of the result Saturates the result in case of underflows or overflows Negates the 32 bit L_v1 with saturation saturate in the case where input is 2147483648 Compares two 32 bit variables L vl and L_v2 and returns the maximum value Rec ITU T G 191 03 2010 9 L min L vl L v2 norm_1 L_v1 L_mult v1 v2 L multo v1 v2 mult v1 v2 mult_r v1 v2 L mac L_v3 vl v2 L mac0 L_ v3 vl v2 L macNs L_v3 vl v2 10 Rec ITU T G 191 03 2010 Compares two 32 bit variables L vl and L_v2 and returns the minimum value Produces the number of left shifts needed to normalize the 32 bit variable L_v1 for positive values on the interval with minimum of 1073741824 and maximum 2147483647 and for negative values on the interva
29. of the ITU T G 711 module g726demo c on the use of the ITU T G 726 module g727demo c on the use of the ITU T G 727 module g722demo c on the use of the ITU T G 722 module g728enc c on the use of the ITU T G 728 floating point encoder g728dec c on the use of the ITU T G 728 floating point decoder g728fpenc c on the use of the ITU T G 728 fixed point encoder g728fpdec c on the use of the ITU T G 728 fixed point decoder rpedemo c on the use of the full rate GSM 06 10 speech codec module sv56demo c on the use of the speech voltmeter module and also the gain loss routine eiddemo c on the use of the error insertion device for bit error insertion and frame erasure eid ev c on the use of the error insertion device for bit error insertion for layered bitstreams which can be used to apply errors to individual layers in layered bitstreams such as ITU T G 718 or ITU T G 729 1 gen patt c on the use of generating bit error pattern files for error insertion in ITU T G 192 compliant serial bitstreams encoded files gen_rate_profile c on the use of fast switching rate profile generation tool firdemo c on the use of the FIR finite impulse response high quality low pass and band pass filters and of the FIR IRS filters associated with the rate change module pemdemo c on the use of the ITU T G 712 standard PCM IIR infinite impulse response filters associated with the rate change module filter c on the use of both the IIR and the
30. opy of the corresponding source code to be distributed under the terms of B 1 and B 2 or Rec ITU T G 191 03 2010 19 e accompany it with the information you received as to where the corresponding source code may be obtained This alternative is allowed only for non commercial distribution and only if you received the module in object code or executable form alone Source code for a work means the preferred form of the work for making modifications to it For an executable file complete source code means all the source code for all modules it contains but as a special exception it need not include source code for modules which are standard libraries that accompany the operating system on which the executable file runs or for standard header files or definitions files that accompany that operating system B 5 You may not copy modify sublicense distribute or transfer the Module except as expressly provided under this General Public License Any attempt otherwise to copy modify sublicense distribute or transfer the Module is void and will automatically terminate your rights to use the Module under this License However parties who have received copies or rights to use copies from you under this General Public License will not have their licenses terminated so long as such parties remain in full compliance B 6 By copying distributing or modifying the Module or any work based on the Module you indicate your acceptance of this lic
31. requency of 16 kHz and returns data at 8 kHz clear state variables needed only if another signal should be processed with the same filter for a parallel form structure deallocate filter memory for a parallel form state variable structure cascade form IIR filtering routine initialization of a cascade form IIR standard PCM filter for data sampled at 8 kHz iir_irs 8khz init iir_casc_1lp 3 to 1 init iir casc_lp 1 to 3 init cascade _iir reset cascade_iir free direct _iir_ kernel iir_dir_dc_removal_ init direct_reset direct _iir free d Error insertion module Name eid c Associated header file eid h Functions included open _eid open_burst_eid reset _burst_eid BER_generator FER _generator_random FER _generator_burst BER insertion FER_module close eid e G 711 module Name g711 c Associated header file g711 h Functions included initialization of a cascade form IIR P 48 IRS filter for data sampled at 8 kHz initialization of a cascade form IIR low pass filter for asynchronization filtering of data and downsampling by a factor of 3 1 initialization of a cascade form IIR low pass filter for asynchronization filtering of data and upsampling by a factor of 3 1 clear state variables needed only if another signal should be processed with the same filter for a cascade form structure deallocate filter memory for a cascade form state variable structure direct
32. s can be used i_mult v1 v2 Multiplies two 16 bit variables v1 and v2 returning a 16 bit value with overflow control rotl vl v2 v3 Rotates the 16 bit variable v1 by 1 bit to the most significant bits Bit 0 of v2 is copied to the least significant bit of the result before it is returned The most significant bit of v1 is copied to the bit 0 of v3 variable rotr vl v2 v3 Rotates the 16 bit variable v1 by 1 bit to the least significant bits Bit 0 of v2 is copied to the most significant bit of the result before it is returned The least significant bit of v1 is copied to the bit 0 of v3 variable L_rotl L_vl v2 v3 Rotates the 32 bit variable L_vl by 1 bit to the most significant bits Bit 0 of v2 is copied to the least significant bit of the result before it is returned The most significant bit of L_vl1 is copied to the bit 0 of v3 variable Rec ITU T G 191 03 2010 13 L rotr L_v1 v2 v3 L_sat L_v1 L_mls L_vl v2 div_s v1 v2 div_1 L_v1 v2 L40_add L40 v1 L40_v2 L40_sub L40 v1 L40_v2 L40_abs L40_v1 L40_shl L40 v1 v2 L40_shr L40 v1 v2 L40_negate L40_ v1 14 Rec ITU T G 191 03 2010 Rotates the 32 bit variable L vl by 1 bit to the least significant bits Bit 0 of v2 is copied to the most significant bit of the result before it is returned The least significant bit of L_v1 is copied to the bit 0 of v3 variable Long 32 bit L vl is set to 2147483647 if an overflow occ
33. t bits and rounds the result It is equivalent to L40_shr L40_v1 v2 except that if v2 is positive and the last shifted out bit is 1 then the shifted result is incremented by 1 without saturation control on 40 bits Any detected overflow on 40 bits will exit execution Arithmetically shifts the 40 bit variable L40 vl by v2 positions to the most significant bits and rounds the result It is equivalent to L40_shl L40_varl v2 except if v2 is negative In this case it does the same as L40 shr r L40_vl v2 L40_set L40_v1 Assigns a 40 bit constant to the returned 40 bit variable Mpy_32_32_ss L_vl L_v2 Multiplies the 2 signed 32 bit values L_vl and L_v2 with L v3 h L v3 1 saturation control on 64 bits The operation is performed in fractional mode When L vl and L v2 are in 1Q31 format the result is produced in 1Q63 format L_v3_h bears the 32 most significant bits while L_v3_1 bears the 32 least significant bits The Basic Operators module is supplemented by two tools one to evaluate program ROM complexity for fixed point code and another to evaluate complexity including program ROM of floating point implementations n l Program ROM estimation tool for fixed point C code Name basop_cnt c Associated header file None Usage basop cnt input c result _file name txt The basop_cnt tool estimates the program ROM of applications written using the ITU T Basic Operator libraries It counts the n
34. ub L_v3 L_mult vl v2 32768 Performs a bit wise AND between the two 16 bit variables vl and v2 Performs a bit wise OR between the two 16 bit variables v1 and v2 Performs a bit wise XOR between the two 16 bit variables vl and v2 Logically shifts left the 16 bit variable v1 by v2 positions if v2 is negative vl is shifted to the least significant bits by v2 positions with insertion of 0 at the most significant bit if v2 is positive v1 is shifted to the most significant bits by v2 positions without saturation control Rec ITU T G 191 03 2010 11 lshr v1 v2 L and L v1 L v2 L or L vl L v2 L xor L vl L v2 L lshl L_v1 v2 L lshr L_v1 v2 extract_h L_v1 extract_1 L_v1 round L_v1 L_deposit_h v1 L_deposit_1 v1 L add c L vl L v2 L sub c L vl L v2 12 Rec ITU T G 191 03 2010 Logically shifts right the 16 bit variable v1 by v2 positions if v2 is positive v1 is shifted to the least significant bits by v2 positions with insertion of 0 at the most significant bit if v2 is negative v1 is shifted to the most significant bits by v2 positions without saturation control Performs a bit wise AND between the two 32 bit variables L_vl and L v2 Performs a bit wise OR between the two 32 bit variables L_vl and L v2 Performs a bit wise XOR between the two 32 bit variables L_vl and L v2 Logically shifts left the 32 bit variable L_v
35. umber of calls to basic operators in the input C source file and also the number of calls to user defined functions The sum of these two numbers gives an estimation of the required PROM n 2 Complexity evaluation tool for floating point C code Name flc c Associated header file flc h Functions included FLC_init Initialize the floating point counters FLC_sub_start Marks the start of a subroutine subsection FLC_sub_end Marks the end of a subroutine subsection FLC_end Computes and prints complexity i e floating point counter results FLC_frame_update Marks the end of a frame processing to keep track of the per frame maxima o Reverberation module Name reverb lib c Associated header file reverb lib h Functions included conv Convolution routine shift Shift elements of a vector for the block based convolution Rec ITU T G 191 03 2010 17 p Bit stream truncation module Name trunc lib c Associated header file trunc lib h Functions included trunc Frame truncation routine q Frequency response calculation module Name fft c Associated header file fft h Functions included rdft Discrete Fourier Transform for real signals genHanning Hanning window generation routine powSpect Power spectrum computation routine 18 Rec ITU T G 191 03 2010 Annex B ITU T software tools General Public License This annex forms an integral part of this Recommendation Terms and conditions B 1 This Lic
36. urred or 2147483648 if an underflow occurred on the most recent L_add_c L_sub c L_macNs or L_msuNs operations The carry and overflow values are binary variables which can be tested and assigned values Performs a multiplication of a 32 bit variable L_vl by a 16 bit variable v2 returning a 32 bit value Produces a result which is the fractional integer division of vl by v2 Values in v1 and v2 must be positive and v2 must be greater than or equal to v1 The result is positive leading bit equal to 0 and truncated to 16 bits If v1 equals v2 then div vl v2 32767 Produces a result which is the fractional integer division of a positive 32 bit variable L_v1 by a positive 16 bit variable v2 The result is positive leading bit equal to 0 and truncated to 16 bits Adds the two 40 bit variables L40 _ v1 and L40_v2 without saturation control on 40 bits Any detected overflow on 40 bits will exit execution Subtracts the two 40 bit variables L40 v2 from L40 vl without saturation control on 40 bits Any detected overflow on 40 bits will exit execution Returns the absolute value of the 40 bit variable L40 v1 without saturation control on 40 bits Any detected overflow on 40 bits will exit execution Arithmetically shifts left the 40 bit variable L40 vl by v2 positions if v2 is negative L40 v1 is shifted to the least significant bits by v2 positions with extension of the sign bit if v2 is positive L40_v1
Download Pdf Manuals
Related Search