diff --git a/apps/encoder.c b/apps/encoder.c index aae2e1e88e1766814bff4987d2b6aab8156d59a8..91932dd8af40b955e75cf9b66aa16eb6ff2348b9 100644 --- a/apps/encoder.c +++ b/apps/encoder.c @@ -1648,18 +1648,17 @@ static void usage_enc( void ) fprintf( stdout, "Options:\n" ); fprintf( stdout, "--------\n" ); fprintf( stdout, "EVS mono is default, for IVAS choose one of the following: -stereo, -ism, -sba, -masa, -mc\n" ); - fprintf( stdout, "-stereo [Mode] : Stereo format, default is unified stereo \n" ); - fprintf( stdout, " optional for Mode: 1: DFT Stereo, 2: TD Stereo, 3: MDCT Stereo\n" ); + fprintf( stdout, "-stereo : Stereo format \n" ); fprintf( stdout, "-ism (+)Ch Files : ISM format \n" ); fprintf( stdout, " where Ch specifies the number of ISMs (1-4)\n" ); - fprintf( stdout, " where positive (+) means extended metadata format is used (including orientation and radius) \n" ); + fprintf( stdout, " where positive (+) indicates extended metadata (only 64 kbps and up) \n" ); fprintf( stdout, " and Files specify input files containing metadata, one file per object\n" ); fprintf( stdout, " (use NULL for no input metadata)\n" ); fprintf( stdout, "-sba +/-Order : Scene Based Audio input format (Ambisonics ACN/SN3D),\n" ); fprintf( stdout, " where Order specifies the Ambisionics order (1-3),\n" ); fprintf( stdout, " where positive (+) means full 3D and negative (-) only 2D/planar components to be coded\n" ); - fprintf( stdout, "-masa Channels File : MASA format \n" ); - fprintf( stdout, " where Channels specifies the number of input/transport channels (1 or 2): \n" ); + fprintf( stdout, "-masa Ch File : MASA format \n" ); + fprintf( stdout, " where Ch specifies the number of input/transport channels (1 or 2): \n" ); fprintf( stdout, " and File specifies input file containing parametric MASA metadata \n" ); fprintf( stdout, "-mc InputConf : Multi-channel format\n" ); fprintf( stdout, " where InputConf specifies the channel configuration: 5_1, 7_1, 5_1_2, 5_1_4, 7_1_4\n" ); diff --git a/readme.txt b/readme.txt index c566c78ea233901d72fc367809458787ddc2e9fa..397de57438376dc40fddf2b0560bad76291488be 100644 --- a/readme.txt +++ b/readme.txt @@ -123,32 +123,33 @@ should have the following structure: |-- Workspace_msvc |-- apps |-- lib_com -....|-- lib_debug + |-- lib_debug |-- lib_dec |-- lib_enc + |-- lib_rend |-- lib_util |-- scripts -....|-- work_in_progress - |-- readme_OSS.txt - `-- readme.txt + |-- tests + |-- readme.txt The package includes a Makefile for gcc, which has been verified on 32-bit Linux systems. The code can be compiled by entering the directory -"c-code" and typing the command: make. The resulting encoder/decoder -executables are named "IVAS_cod" and "IVAS_dec". Both reside in the c-code -directory. +"c-code" and typing the command: make. The resulting encoder/decoder/renderer +executables are named "IVAS_cod", "IVAS_dec", and "IVAS_rend". All reside +in the c-code directory. The package also includes a solution-file for Microsoft Visual Studio 2017 (x86). To compile the code, please open "Workspace_msvc\Workspace_msvc.sln" and build "encoder" for the encoder and "decoder" for the decoder executable. The resulting -encoder/decoder executables are named "IVAS_cod.exe" and "IVAS_dec.exe". Both reside -in the c-code directory. +encoder/decoder/renderer executables are named "IVAS_cod.exe", "IVAS_dec.exe", +and "IVAS_rend.exe". All reside in the c-code directory. RUNNING THE SOFTWARE ==================== The usage of the "IVAS_cod" program is as follows: +-------------------------------------------------- Usage: IVAS_cod.exe [Options] R Fs input_file bitstream_file @@ -161,9 +162,9 @@ R : Bitrate in bps, for AMR-WB IO modes R = (6600, 8850, 12650, 14250, 15850, 18250, 19850, 23050, 23850) for IVAS stereo R = (13200, 16400, 24400, 32000, 48000, 64000, 80000, - 96000, 128000, 160000, 192000, 256000) - for IVAS ISM R = 13200 for 1 ISM, 16400 for 1 ISM and 2 ISM, - (24400, 32000, 48000, 64000, 80000, 96000, 128000) + 96000, 128000, 160000, 192000, 256000) + for IVAS ISM R = 13200 for 1 ISM, 16400 for 1 ISM and 2 ISM, + (24400, 32000, 48000, 64000, 80000, 96000,128000) for 2 ISM, 3 ISM and 4 ISM also 160000, 192000, 256000 for 3 ISM and 4 ISM also 384000 for 4 ISM also 512000 @@ -179,17 +180,17 @@ bitstream_file : Output bitstream filename Options: -------- EVS mono is default, for IVAS choose one of the following: -stereo, -ism, -sba, -masa, -mc --stereo [Mode] : Stereo format, default is unified stereo - optional for Mode: 1: DFT Stereo, 2: TD Stereo, 3: MDCT Stereo --ism Channels Files : ISM format - where Channels specifies the number of ISMs (1-4) +-stereo : Stereo format +-ism [+]Ch Files : ISM format + where Ch specifies the number of ISMs (1-4) + where positive (+) indicates extended metadata (only 64 kbps and up) and Files specify input files containing metadata, one file per object (use NULL for no input metadata) -sba +/-Order : Scene Based Audio input format (Ambisonics ACN/SN3D), where Order specifies the Ambisionics order (1-3), where positive (+) means full 3D and negative (-) only 2D/planar components to be coded --masa Channels File : MASA format - where Channels specifies the number of input/transport channels (1 or 2): +-masa Ch File : MASA format + where Ch specifies the number of input/transport channels (1 or 2): and File specifies input file containing parametric MASA metadata -mc InputConf : Multi-channel format where InputConf specifies the channel configuration: 5_1, 7_1, 5_1_2, 5_1_4, 7_1_4 @@ -213,13 +214,13 @@ EVS mono is default, for IVAS choose one of the following: -stereo, -ism, -sba, -mime : Mime output bitstream file format The encoder produces TS26.445 Annex.2.6 Mime Storage Format, (not RFC4867 Mime Format). default output bitstream file format is G.192 --agc op : SBA Adaptive gain control, op = (0, 1), by default op is 0 or deactivated -bypass mode : SBA PCA by-pass, mode = (1, 2), 1 = PCA off, 2 = signal adaptive, default is 1 -q : Quiet mode, no frame counters default is deactivated The usage of the "IVAS_dec" program is as follows: +-------------------------------------------------- Usage for EVS: IVAS_dec.exe [Options] Fs bitstream_file output_file Usage for IVAS: IVAS_dec.exe [Options] OutputConf Fs bitstream_file output_file @@ -268,13 +269,12 @@ Options: left or l or 1->left, right or r or -1->right, center or c or 0->middle -q : Quiet mode, no frame counter default is deactivated --FEC X : Insert frame erasures, X = 0-10 is the percentage - of erased frames, or X may be the name of binary file or - file with G192 headers indicating GOOD FRAME or BAD FRAME - containing FEC pattern (short values of 0 (good) or 1 (bad)) - default is OFF, if this option is not used --force R : Force specific binaural rendering mode, R = (TDREND, CLDFBREND), - + + +The usage of the "IVAS_rend" program is as follows: +--------------------------------------------------- + +TBD MULTICHANNEL LOUDSPEAKER INPUT / OUTPUT CONFIGURATIONS @@ -359,7 +359,7 @@ stv2MASA2TC48c.wav - 2 channels (2 MASA transport channel), 48000 Hz, 48000 Hz, For the MASA operation modes, in addition the following metadata files -are required: +located in /scripts/testv/ folder are required: stv1MASA1TC48c.met stv1MASA1TC48n.met @@ -375,7 +375,7 @@ available at https://www.3gpp.org/ftp/TSG_SA/WG4_CODEC/TSGS4_118-e/Docs/S4-220443.zip For the ISM operation modes, in addition the following metadata files -are required: +located at /scripts/testv/ folder are required: stvISM1.csv stvISM2.csv @@ -384,7 +384,7 @@ stvISM4.csv These are comma separated files (csv) which indicate the per object position in the format: -frame azimuth, elevation, distance (unit circle), spread, gain +frame azimuth, elevation, radius, spread, gain, yaw, pitch with the following meaning: | Parameter | format, value range | meaning @@ -393,12 +393,23 @@ with the following meaning: ----------------------------------------------------------------------------------- | elevation | float, [-90,90] | elevation; positive indicates up ----------------------------------------------------------------------------------- -| distance | float, tbd | distance; default: 1 +| radius | float, [0, 15.75] | radius (extended metadata) ----------------------------------------------------------------------------------- | spread | float, [0,360] | spread in angles from 0...360 deg; default: 0 ----------------------------------------------------------------------------------- | gain | float, [0,1] | gain; default: 1 ----------------------------------------------------------------------------------- +| yaw | float, [-180,180[ | yaw (extended metadata); positive indicates left +----------------------------------------------------------------------------------- +| pitch | float, [-90,90] | pitch (extended metadata); positive indicates up +----------------------------------------------------------------------------------- + +For the HRTF filter File option, external HRTF filter Files are available in folder +/scripts/binauralRenderer_interface/binaural_renderers_hrtf_data : + +ivas_binaural_16kHz.bin +ivas_binaural_32kHz.bin +ivas_binaural_48kHz.bin For the Head rotation operation modes, external trajectory files are available: @@ -409,6 +420,9 @@ headrot_case01_3000_q.csv headrot_case02_3000_q.csv headrot_case03_3000_q.csv +For Reference vector specified by external trajectory file, example files are available at +/scripts/trajectories folder. + For the Renderer configuration option operation modes, external configuration files are available: @@ -421,5 +435,6 @@ config_renderer.cfg ================== Additional scripts for item generation and codec testing are available -in the directory scripts. Please refer to scripts/README.md for -additional documentation. +in the directories scripts and tests. Please refer to scripts/README.md, resp. +tests/README.md for additional documentation. +