Audio engineer compressor techniques and applications

Audio Engineer-Compressor

Audio compressor

An audio compressor is a signal processor that works with sound. It can be a piece of hardware, software, or a combination of software and hardware. Sound is an analog signal that causes air molecule to vibrate, and propagate the energy until it reaches the ears. The advent of electrical and electronic devices meant that sound could be transmitted as voltage level signals in telephones and radio at the beginning of the twentieth century. There was also a need to store information such as music.

The advent of storage of sound information and electrical transmission of sound necessitated the processing of the sound signal. Before audio processing, radio stations had to avoid over-modulation through "manual gain riding." A person, with a hand on a knob, anticipated the signal peaks. The process was not efficient and failure to prevent over modulation meant overloading the transmitter and dropping off-air. In 1935, Al Towne developed the first audio processor to have an intelligent compressor (Hudson 23). However, it appeared in the market about ten years later.

The advances in technology in the twentieth century brought about complex and highly developed audio compressors. The vacuum tubes and the transistors had a significant impact. The leading company in this front was General Electric. However, it was not until the 1960s that a true audio processer emerged: The Audio Pilot

The foregoing discussion deals with the early application of audio compression. However, the other aspect of audio compression is the reduction of the actual data for transmission or storage. The earliest known audio compression occurred when in 1939, Dudley introduced the VOCODER. This reduced the amount of bandwidth needed to transmit speech over a telephone line. This area of audio compression has developed extensively, especially with the development of computers in the late 1970s.

Audio compression can be digital or analog. The current advances in technology mean that most information must at one time or another be converted in to a digital format. I will, therefore, discuss the digital aspect of audio compression to a larger extent. The main goals in audio compression are to reduce bandwidth, reduce storage space, ensure a low implementation complexity, and make the decoded signal as close as possible to the original signal (Gaydecki 1592).

The capacity of an audio channel limits the quality of audio signals. The bandwidth and the ratio of signal level to noise in a transmission channel limit analog signals. On the other hand, the sampling ratio and the sample word-length limit digital signals. The cost and power necessary in the signal transmission, as well as the cost of data storage, are key driving factors behind audio compression technology. Reduction of the bit rate translates to a smaller bandwidth requirement, and low transmitter power.

Compression reduces the amount of storage medium required in direct proportion to the compression factor. This is especially necessary in archiving as it reduces the size and cost of the Compression lowers the bit rate of a signal. The ratio between the original bit rate and compressed bit rate is the compression factor. Compression can, therefore, be termed as bit rate reduction. Compression also may reduce the size and weight of recorder, as well as relax the mechanical tolerance of audio devices.

In cinema, audio compression maximizes the number of audio channels. Consumer devices such as games, portable digital devices, and music players also have audio compression technology. Further, these devices access the internet and may play or use internet content. The sheer numbers involve necessitate the use of audio compression in the storage and archiving of this online content.

Implementing audio compression requires a good understanding of the human ear. Only a certain portion of sound is useful to the ear. This is the perceptual entropy. All other sound is redundant. The ear distinguishes certain critical frequency bands. The result is an effective Q. factor responsible for auditory masking. A detailed modeling of how masking occurs in the ear is essential to the design of audio compressors. It is vital not to exceed the masking threshold so that compression is not audible.