10. March 2021

The latency (time) of an audio system refers to the time difference from the moment a signal is fed into the system to the moment it appears at the output. Depending on the application, such a delay can have various effects. Usually, the aim is to achieve the lowest possible latency - but there are also applications where latency is used deliberately. This article provides an overview of the topic and describes methods for measuring latency in audio systems.

Latency during calls

Online conferences have become part of everyday life in the working world. Pretty much everyone has thus encountered this problem: audio communication is severely delayed, resulting in the participants talking over each other. This makes communication difficult and frustrating.

Cause and Effect of Latency in Audio Systems

There are three main causes of latency.

In the case of the "online conference" example, delays are caused, for example, by the buffering of the transmitted data. Error correction algorithms also contribute to this. In extreme cases, this can result in latency times of several seconds.

Another cause is the physical running times of transmission paths, particularly through air. At large events, for example, it takes a noticeable time, and manifests in a significant drop in sound level, when sound travels through the air from a loudspeaker to the audience positioned up to several hundred meters away. To compensate for the resulting reduction in sound level, so-called support speakers can be positioned further back in the auditorium. If the music signal were fed into the loudspeakers next to the stage and the support loudspeakers at the same time, the audience in the back of the room would hear the sound from the support speakers first and the sound from the main loudspeakers with a delay. The audience would thus perceive the support speakers as the main sound source and not the stage as desired. (see Haas Effect) For this reason, the support speakers are fed a delayed signal. To put it simply, this delay corresponds to the time it takes for the sound to travel from the main loudspeaker to the location of the supporting loudspeakers.

In digital signal processing, latency occurs in A/D and D/A conversion, as well as in various algorithms (effects, error correction, etc.). Depending on the application, a maximum latency must not be exceeded.  If, for example, a movie is viewed using wireless headphones, a latency of 150 ms to 200 ms can already be disturbing, depending on personal perception. In dialogue, lip movement is no longer synchronized.

With digital InEar-Monitor (IEM) systems, the requirements are much stricter. The artist hears his singing indirectly through the structure-borne sound of the skull. In addition, he hears his voice through the IEM system. Here, only a few milliseconds of latency are acceptable.

Latency Measurement

To determine acoustic delay times or the distance to a loudspeaker, the XL2 Audio and Acoustic Analyser is suitable in combination with the MR-PRO Audio Signal Generator. This allows delay times for support loudspeakers to be measured quickly and precisely.

The latency of electrical and acoustic systems, the effect of algorithms on latency, as well as the delay time of audio transmission channels can be determined with the FX100 Audio Analyser. The required test signal comes from the built-in generator.

Eager to learn more?

Register for our free webinar "Latency Fundamentals and Measurement". Besides an insight into the theory of different measurement methods, practical examples of different applications will be shown.