Normalization (audio)

from Wikipedia, the free encyclopedia

In audio technology, normalization is the process of increasing or decreasing the amplitude of analog or digital audio material so that it lies within a specified range. There are essentially two reasons why normalization takes place - on the one hand, technical framework conditions for the transmission or storage of the audio material, and on the other hand, the adjustment of the loudness ( volume perceived by humans ) of the material.

A technical object can be present, a signal up to the maximum possible amplitude auszusteuern to (after reduction of the word length) provides a good signal-to-noise ratio and low quantization error to obtain. One application from the area of ​​loudness is to adjust the same to a specific target value or setpoint , as is done e.g. B. in the recommendations EBU-R 128 , ITU BS.1770 or Replay Gain is described.

This is very often used e.g. B. at radio stations for pop music, the normalization to adjust the loudness impression between different titles, so that the listener is not tempted or compelled to readjust the volume on the receiver during quieter titles. This is often combined with a reduction in the dynamic range by means of an audio compressor , which automatically reduces the loudness differences within a piece of music. This is described in more detail in the articles FM Sound Processing and Automatic Gain Control.

In software for audio signal processing there is usually the possibility to have these mathematical processes (see also standardization ) carried out automatically:

  • Adjustment according to maximum amplitude: The audio material is searched for the point at which the signal has the greatest absolute value. Then the entire material is amplified or weakened exactly to the extent that this value becomes the desired (permitted) maximum value.
  • RMS normalization (Root Mean Square): This method does not take into account the maximum value but the effective value (RMS) of the material.
  • Loudness normalization: With this method, the loudness of audio material is calculated according to a specific algorithm and changed to a specific target value (Target Level −23 LUFS for EBU-R 128). This type of loudness adjustment makes it possible to adjust the loudness according to the human perception of loudness in order to e.g. B. adapt several pieces of music to each other or to avoid loudness jumps at program transitions in radio and television.

literature

  • Gustav Büscher, A. Wiegemann: Little ABC of electroacoustics. 6th edition, Franzis Verlag, Munich 1972, ISBN 3-7723-0296-3
  • EBU Technical Recommendation R 128 'Loudness normalization and permitted maximum level of audio signals' (2011)
  • EBU Tech Doc 3342 'Loudness Range: A parameter to supplement loudness normalization in accordance with EBU R 128' (2011)
  • Roland Enders: The home recording manual. 3rd edition, Carstensen Verlag, Munich 2003, ISBN 3-910098-25-8
  • ITU-R BS.1770 'Algorithms to measure audio programs loudness and true-peak audio level'

Web links