CJOpus
Emscripten (C++/JavaScript) bindings for libopus
A module to encode PCM data to and decode PCM data from Opus.
Using
The lib folder includes a pre-compiled version. Requiring the Opus.js entrypoint file should be enough.
OpusEncoder
//Tries to mimic `node-opus` in syntax.
var encoder = new OpusEncoder( 48000, 2 );
OpusEncoder#encode(Buffer/TypedArray/Array)
var PCM = getPCMDataSomehow();
var encoded = encoder.encode( PCM );
OpusEncoder#decode(Buffer/TypedArray/Array)
var OPUS = getOPUSDataSomehow();
var decoded = encoder.decode( OPUS );
OpusEncoder#encodeUnsafe / OpusEncoder#decodeUnsafe
These two methods work just like the safe versions, however they don't do any error checking and they use .subarray() instead of .slice(). Because of this, it's a bit faster. It's relatively safe to use, just know that if an encoding error happens, it will return an empty Uint8Array (encode) or Int16Array (decode).
OpusEncoder#destroy()
encoder.destroy();
Since this is an Emscripten module (that also has to malloc), the memory also needs to be freed manually when finished. It's a small program, perhaps under 30KB of dynamically allocated memory, but that can add up.
Building
The required packages for building (under Debian-based distros) are:
- git
- build-essential
- libtool
$ git clone --recursive https://github.com/izy521/CJOpus.git
$ cd CJOpus
$ make
Source
| 1 | # CJOpus |
| 2 | Emscripten (C++/JavaScript) bindings for libopus |
| 3 | |
| 4 | A module to encode PCM data to and decode PCM data from Opus. |
| 5 | |
| 6 | ## Using |
| 7 | |
| 8 | The `lib` folder includes a pre-compiled version. Requiring the `Opus.js` entrypoint file should be enough. |
| 9 | |
| 10 | ### OpusEncoder |
| 11 | |
| 12 | ```js |
| 13 | //Tries to mimic `node-opus` in syntax. |
| 14 | var encoder = new OpusEncoder( 48000, 2 ); |
| 15 | ``` |
| 16 | |
| 17 | ### OpusEncoder#encode(Buffer/TypedArray/Array) |
| 18 | |
| 19 | ```js |
| 20 | var PCM = getPCMDataSomehow(); |
| 21 | var encoded = encoder.encode( PCM ); |
| 22 | ``` |
| 23 | |
| 24 | ### OpusEncoder#decode(Buffer/TypedArray/Array) |
| 25 | |
| 26 | ```js |
| 27 | var OPUS = getOPUSDataSomehow(); |
| 28 | var decoded = encoder.decode( OPUS ); |
| 29 | ``` |
| 30 | |
| 31 | ### OpusEncoder#encodeUnsafe / OpusEncoder#decodeUnsafe |
| 32 | |
| 33 | These two methods work just like the safe versions, however they don't do any error checking and they use `.subarray()` instead of `.slice()`. Because of this, it's a bit faster. It's relatively safe to use, just know that if an encoding error happens, it will return an empty `Uint8Array` (encode) or `Int16Array` (decode). |
| 34 | |
| 35 | ### OpusEncoder#destroy() |
| 36 | |
| 37 | ```js |
| 38 | encoder.destroy(); |
| 39 | ``` |
| 40 | |
| 41 | Since this is an Emscripten module (that also has to malloc), the memory also needs to be freed manually when finished. It's a small program, perhaps under 30KB of dynamically allocated memory, but that can add up. |
| 42 | |
| 43 | ## Building |
| 44 | |
| 45 | The required packages for building (under Debian-based distros) are: |
| 46 | |
| 47 | * git |
| 48 | * build-essential |
| 49 | * libtool |
| 50 | |
| 51 | ```bash |
| 52 | $ git clone --recursive https://github.com/izy521/CJOpus.git |
| 53 | $ cd CJOpus |
| 54 | $ make |
| 55 | ``` |