Basics of digital audio:
Before the computer can record, manipulate, and reproduce sound, sound must be transformed from an audible analog form to a computer-acceptable digital form, using a process called analog-to-digital conversion (ADC). Once the sound data has been stored as bytes in the computer, the power of the computer’s CPU can be used to transform this sound in thousands of ways. Finally, when you are ready to listen to the result, the digital-to-analog conversion (DAC) process transforms the sound bytes back into an analog electrical signal from the speakers.
Sampling: Analog to Digital Conversion
Given an analog signal, discrete values of its amplitude are taken at small time intervals, obviously the more reliable the reproduction the more samples per second are taken. These obtained values are assigned a digital value that the computer can understand and process as required. We can use 8 or 16 bit words, thus obtaining 256 or 65536 different combinations and obtaining higher resolution.
SAMPLE FREQUENCY: According to the Nyquist theorem, it is possible to accurately repeat a waveform if the sampling frequency is at least twice the frequency of the component with the highest frequency. The highest frequency that the human ear can perceive is close to 20 kHz, so the 44.1 kHz sampling rate of sound cards is more than enough. This value is the one used today by CD audio players.
SAMPLE SIZE: The sample size controls the dynamic range that can be recorded. For example, 8-bit samples limit the dynamic range to 256 steps (50 dB range). In contrast, a 16-bit sample has a dynamic range of 65,536 steps (90 dB range) a substantial improvement. The human ear perceives a whole world of differences between these two sample sizes. Ears are more sensitive to detecting differences in pitch than intensity, but are even more sensitive to the strength of sound.
From the previous processes we can get an audio file, such as (and since it is the best known), a WAV audio file. It is the own format of Windows. They can be 8 or 16 bit with sampling rates of 11,025 kHz, 22.05 kHz, or 44.1 kHz and generally have good sound quality.
Digital audio compression
It could be assumed that all you have to do to get good sound is to record at the 44.1 kHz speed limit with 16-bit (2-byte) samples. The only problem that appears if recording in stereo, sampling simultaneously on the left and right channels at 44.1 kHz, a one minute sound sample needs a 10.58MB storage space. This involves using large disk spaces to store these sound files. Many compressed file formats (codecs) have been developed that enable high-quality recording without the need for so much disk space.
Most common audio formats:
With the simple objective of listing a series of codecs used by different operating systems to perform audio compression. Later, a more complete description of the most used is made: MP3.
Therefore, some of the most used are:
Advanced Audio Coding (AAC): used by Apple computers. More efficient than MP3.
Audio for Unix (AU): Acoustic standard for the JAVA programming language.
Windows Media Audio (WMA)
Ogg Vorbis: It is free, open and not patented.
Atrac: compression and playback technology for minidisc.
The codec par excellence: the MP3
Its origin and current
The abbreviations MP3 respond to the abbreviation of MPEG (Moving Picture Expert Group) 1 Layer 3, which is a perceptual coding algorithm. This among others was developed by the Moving Picture Expert Group (MPEG) (http://www.cselt.it/mpeg/) together with the Fraunhofer Institute of Technology (http://www.ipa.fhg.de/english/ ).
Moving Picture Expert Group is an ISO / IEC research committee. MPEG is in charge of the international development of compression, decompression, processing and encoded rendering standards for movies, audio and the combination of both. It is a non-profit institution created in 1988, which brings together 300 experts from 20 countries three times a year.
