Lossy Formats


To introduce the subject, first, we need to define compression algorithms as a tool to reduce the memory required to hold information. To transmit information, audio in this case, between users, sometimes it’s necessary compress the file with the aim to reduce its size and thus make a faster and optimal transfer. Continue reading




A codec transforms a file with a data stream or signal. Codecs can encode the stream or signal (often for transmission, storage or encryption) and retrieve or decrypt the same way for reproduction or manipulation in an appropriate format for these operations.

An audio codec is a codec that includes a set of algorithms to encode and decode the auditory data, which means reducing the number of bits occupied by the audio file.

Most codecs are lossy, to achieve the smallest possible size of the target file. There are also lossless codecs, but in most practical applications, for an almost imperceptible increase in quality is not worth a considerable increase in data size. The exception is if the data undergo other treatments in the future. Repeated lossy encoding ultimately hurt the quality too.

– Lossy audio codecs

In audio limitations of the human auditory system are used to compress audio data. Different methods are used to remove inaudible data:

  • Transformed into the frequency domain and removing inaudible frequency (below 20Hz and above 20kHz, or masked by other frequencies higher power).
  • Dividing the signal into frequency subbands that approximate critical bands and quantifying each subband based on a noise detection threshold in the subband. Thus the audio signal is analyzed and the amount of noise (negligible loss) that can be input to each frequency range, that is, the masking threshold is calculated.
  • Prediction data.
  • Characterization of the voice from a parameterization.
  • The compression level can be controlled and depends on the quality to be obtained, the file size, the bandwidth of the network, the compression …

– Lossless audio codecs

These codecs use lossless compression to minimize the size of the data stream. To perform this coding algorithms based on eliminating redundancy from the audio signal is used, and therefore the degree of predictability of information. If the signal has repetitive patterns, it is redundant and therefore easy to predict.

The main objective of the audio codecs is to reduce the amount of digital data required to reproduce an audio signal. It is particularly useful for compressing data that occupy the minimum possible size when stored, and if it has to transmit network can be sent as quickly as possible. Thus, the main applications are:

  • Storage: Useful in Multimedia devices of audio playback or stored, as will be better greater compression coding and thus also the space required for storage.
  • Transmission: useful when you must send the information by any network since a lower bit rate will be faster shipping.


When we use a codec we store the information in a Format. An audio file format is a multimedia container that saves an audio recording. There are two main kinds of audio formats:

– Uncompressed:

PCM: It is not a file type or format. It is a technique of transforming analog audio to digital without any compression. Therefore we don’t see the pcm audio extension. PCM contain all the information that came out of the analog to digital converter, with no default and therefore have the best quality. We work with PCM when digitizing, but always keep in files with any of these extensions:

  • WAV (Wave, Wave in English): It is the format of uncompressed digital audio over used. It belongs to Microsoft / IBM.
  • AIFF (Audio Interchange File Format): It is similar to but for WAV or Apple Macintosh computers MAC.
  • CDA: These are audio tracks recorded on compact disc that also use the PCM system.

All uncompressed files are large. Approximately 10 Mb per minute of audio. These are the formats used to store audio professionally because the quality is very good. But when we don’t need such quality and are short on space, it’s time to use file compression.

– Compressed. There are tho kinds of compressed audio formats:

  • Lossy: sound files “lossy” are those that use a lossy compression algorithm, that is a compression type that represents information, but trying to use to do a smaller amount of information. This makes it impossible to reconstruct exactly the original file information.
  1. MP3 or MPEG-1 Audio Layer 3: This is a standard digital audio format compressed lossy data loss mp3 format is not audible to the human ear, so do not distinguish the difference between an uncompressed audio file and an mp3 file.
    Also get an mp3 file to reduce file size without affecting sound quality, about 1 minute mp3 audio occupies 1 MB with a quality almost equal to the quality of Cd.
  2. ACC or Advanced Audio Coding: A standard digital audio format as an extension of MPEG-2 compressed with loss and offers higher quality than MP3 and is more stable for the same number and size Kbps. Compression is based on the same principles as MP3 compression, with the difference that provides the possibility of using sampling rates range from 8 Hz to 96 KHz. The encoding method automatically adapts the number of Kbps (bit rate) required depending on the complexity of the audio stream at all times.
  3. WMA or Windows Media Audio is an audio compression format with a loss but there is also this format with lossless compression. It is developed primarily for commercial purposes built into Windows, Windows Media Player player. Is below the level of the above formats.
  • Lossless: sound files “lossless” are those not using or compression methods represent the information without trying to use less of the original information. Enable an accurate reconstruction of the original information.

  1. FLAC or Free Lossless Audio Codec: is able to reduce the size of a sound file from half to three quarters of the original size. The FLAC format is commonly used to sell music online, and as an alternative to MP3 compartila when you want to reduce the size trendría WAV PCM file without losing quality because this type of compression we can reconstruct the original data the file.
  2. MP3 HD: It is the variant of the mp3 format, but in high quality and without loss of information. It has a sampling rate of 44.1 kHz and a resolution of 16 bit stereo. Their quality is similar to that of an uncompressed .wav file. It is compatible with most MP3 players. Its extension is * .mp3hd




Media Refference: