diff options
author | Jean-Marc Valin <jean-marc.valin@octasic.com> | 2010-07-08 13:40:07 -0400 |
---|---|---|
committer | Jean-Marc Valin <jean-marc.valin@octasic.com> | 2010-07-08 13:40:07 -0400 |
commit | 4468c75adbefed247ae592e42549da544f468277 (patch) | |
tree | e95a37dac0ef4604085330f491b109cc4c5a3248 /README | |
parent | ed8e1b49370e39b776930918e5d5f7d6408c1a3c (diff) | |
download | opus-4468c75adbefed247ae592e42549da544f468277.tar.gz |
draft update, instructions
Diffstat (limited to 'README')
-rw-r--r-- | README | 40 |
1 files changed, 28 insertions, 12 deletions
@@ -1,20 +1,36 @@ -This is a prototype codec and for now it doesn't do anything useful. To get -the celt and silk submodules, you need to do: +This is a prototype codec and for now it has limited functionality. -% git submodule init -% git submodule update +To build from a distribution tarball, you only need to do the following: -Compiling is currently tricky. You first need to configure the CELT directory: +% ./configure --disable-shared +% make + +To build from the git repository, the following steps are necessary + +0) Clone the repository: + +% git clone git://git.xiph.org/users/jm/ietfcodec.git +% cd ietfcodec + +1) Get the celt and silk submodules: + +% git submodule update --init + +2) Compiling -% cd celt % ./autogen.sh % ./configure --disable-shared -% cd .. +% make -Then you can use the top-level Makefile by typing: -% make +Once you have compiled the codec, there will be a test_hybrid executable in +the src/ directory. This can be in the following way: + +% ./test_hybrid 48000 1 960 80 input.sw output.sw -If you're lucky, then everything works. Most likely there will be an error in -one of the paths that are hardcoded into one of the makefiles. This should -improve in the future. +The arguments are: +1) The sampling rate (only 48000 supported for now) +2) The number of channels (only mono supported for now) +3) The frame size in samples (480 and 960 supported) +4) Input PCM file (16-bit, machine endian) +5) Output PCM file after encoding and decoding (16-bit machine endian) |