{"id":4730,"date":"2022-05-06T08:49:30","date_gmt":"2022-05-06T06:49:30","guid":{"rendered":"https:\/\/www.pschatzmann.ch\/home\/?p=4730"},"modified":"2023-04-28T16:44:10","modified_gmt":"2023-04-28T14:44:10","slug":"audio-streaming-the-opus-codec","status":"publish","type":"post","link":"https:\/\/www.pschatzmann.ch\/home\/2022\/05\/06\/audio-streaming-the-opus-codec\/","title":{"rendered":"Audio Streaming – The Opus Codec"},"content":{"rendered":"

Codecs play an important role in audio streaming, as we have seen in one of my last blogs<\/a>.<\/p>\n

I have spent some time to make the Opus Codec<\/strong> available as Arduino Library<\/a> and I am providing some simple integration from my Arduino Audio Tools<\/a> library.<\/p>\n

Opus is a codec for interactive speech<\/strong> and audio<\/strong> transmission over the Internet.<\/p>\n

Opus can handle a wide range of interactive audio applications, including Voice over IP, videoconferencing, in-game chat, and even remote live music performances. It can scale from low bit-rate narrowband speech to very high quality stereo music.<\/p>\n

Opus packets are not self-delimiting, but are designed to be used inside a container<\/strong> of some sort which supplies the decoder with each packet’s length. Opus was originally specified for encapsulation in Ogg containers.<\/p>\n

Nevertheless we can test the codec w\/o container.<\/p>\n

Arduino Sketch<\/h2>\n

We can test the codec e.g. with the following Sketch which makes sure that the encoded packets are subsequently decoded again:<\/p>\n

#include \"AudioTools.h\"\n#include \"AudioCodecs\/CodecOpus.h\"\n\nint sample_rate = 24000;\nint channels = 2;  \/\/ The stream will have 2 channels\n\nSineWaveGenerator<int16_t> sineWave( 32000);  \/\/ SoundGenerator max amplitude of 32000\nGeneratedSoundStream<int16_t> sound( sineWave); \/\/ Stream generated from sine wave\nMeasuringStream out; \nOpusAudioEncoder enc;\nEncodedAudioStream decoder(&out, new OpusAudioDecoder()); \/\/ encode and write \nEncodedAudioStream encoder(&decoder, &enc); \/\/ encode and write \nStreamCopy copier(encoder, sound);     \n\nvoid setup() {\n  Serial.begin(115200);\n  AudioLogger::instance().begin(Serial, AudioLogger::Info);\n\n  \/\/ Setup sine wave\n  auto cfgs = sineWave.defaultConfig();\n  cfgs.sample_rate = sample_rate;\n  cfgs.channels = channels;\n  cfgs.bits_per_sample = 16;\n  sineWave.begin(cfgs, N_B4);\n\n  \/\/ Opus decoder needs to know the audio info\n  decoder.begin(cfgs);\n\n  \/\/ configure and start encoder\n  enc.config().application = OPUS_APPLICATION_VOIP;\n  enc.config().complexity = 5;\n  encoder.begin(cfgs);\n\n  Serial.println(\"Test started...\");\n}\n\nvoid loop() { \n  copier.copy();\n}\n\n<\/code><\/pre>\n
Performance Overview<\/h2>\n
Please note that the values of the valid supported sample rates<\/strong> is limited to 8000,12000,16000 ,24000,48000 and there are quite a few encoder settings<\/a> that we can play with, but the most important ones are:<\/p>\n