Codecs are encoders / decoders that digitally encode and decode data or signals to send it over a data network. In VoIP telephony, these play a particularly important role, as they determine how and with what quality the audio and video data is transported within a network.
A general distinction is made between compressed and uncompressed codecs.
In the case of codec compression, a distinction is also made between lossy and lossy compression of multimedia data. With a codec, there is always a trade-off between available bandwidth, the computing power to compress the data, and the quality of voice in both video and audio streaming. Wideband codecs allow very high quality voice transmission. Narrowband codecs allow low-bandwidth voice transmission, but call quality suffers noticeably. In general, a bandwidth of at least 100 kBit / s in both directions is recommended.
With lossless codecs, all audio or video information contained in the original data is preserved in its original quality. With lossy codecs, which often result in a loss of quality, higher compression can be achieved to require less bandwidth during transmission.
If, for example, the audio quality is too low or there are jerks and cuts during the call, it may make sense to change the codec. A common source of poor audio quality with Voice over IP is that the required bandwidth for the codec is not available from the Internet connection. The following codecs are currently used as standard formats:
Audio codecs
GSM: 13 kbit / s voice data rate (full rate method), 20 ms frames
iLBC – 15 kbps, 20 ms voice frames: 13.3 kbps, 30 ms frames
ITU G.711 – 64 kbit / s, based on samples. Also known as the A-law / µ-law method.
ITU G.722 – 48/56/64 Kbit / s
ITU G.723.1 – 5.3 / 6.3 kbit / s, 30 ms frames
ITU G.726 – 16/24/32/40 Kbit / s
ITU G.728 – 16 kbit / s
ITU G.729 – 8 kbit / s, 10 ms frames
Speex: 2.15 to 44.2 kbps
LPC10: 2.5 kbps
DoD CELP – 4.8 kbps
Average Opinion Score (MOS) is a measurement method that allows you to compare the voice quality of different codecs with each other. A group determines how close a codec is to the original human language. The scale ranges from 1 (poor) to 5 (excellent). Values greater than 4 correspond to the voice quality in the ISDN network.
Video codecs
VP8: Free codec to compress HD video (also for streaming)
H.264 / MPEG-4 Part 10 or AVC (Advanced Video Coding) – Very popular paid codec for compressing HD video
All standards are optimized for low data transfer speeds and low video shake. Data transfer rates are not defined.
Codec
In electronics and telecommunications, the term codec is used for integrated circuits or chips that perform data conversion. In this context, the term is an acronym for encoder / decoder. This type of codec combines functions for analog-to-digital conversion with functions for digital-to-analog conversion on a single chip.
In practice, compression is used primarily to optimize the storage space requirements of a large file, such as a video. Compression is achieved by eliminating redundancies in the data.
Compression works for any type of file, including text, program, image, audio, video, or virtual reality (VR) files. The compression rate achieved depends on several factors. In some cases, compression can reduce the size of a file by a factor of 100 or more. For example, a video 30 terabytes in size can be reduced to three terabytes (or less).
Compression achieves the greatest benefit in transmitting data. Uncompressed files like the video file just mentioned are very large and therefore take a long time to upload and download. The compressed file, on the other hand, can generally be downloaded in a few seconds or minutes. In order for the video to be seen again after downloading, a decompression algorithm has to “undo” the compression.
There are numerous standard codec systems. Some are used primarily to minimize file transfer times and are implemented over the Internet. Others are designed to maximize the amount of data to be stored on a storage medium.
