Take the 2-minute tour ×
Signal Processing Stack Exchange is a question and answer site for practitioners of the art and science of signal, image and video processing. It's 100% free, no registration required.

What kind of compression artifacts would typically be most noticeable to a non-audiophile listener in a low-bit rate mp3 (MPEG-2 Layer III) compressed audio file?

What kind of audio signal might one synthesize that would make these artifacts maximally audible between "low" and "high" bit rate mp3 compressed files (using commonly available encoders)?

share|improve this question
    
I think anything with wideband noise will start to sound "warbly" at low bandwidths. Noise is random and therefore not easy to compress. Tones are simple and sparse in the frequency domain and easy to compress. –  endolith Oct 28 '12 at 15:05

1 Answer 1

up vote 5 down vote accepted

Pre-echo on very short and percussive sounds - for example crisp solo drums, or better, electronic drums. The TR-808 rimshot and clave are very short and impulsive sounds, about 5ms long (for the rimshot), and at less than 160kbps their attacks get "swallowed". At less than 96kbps some notes appear to be duplicated or out of place rhythmically because of pre-echo. Worse, each note in a repeated pattern will sound different depending on how it falls within the analysis window. Here is an example.

Try solo vocals too. I think casual listeners are better at detecting anomalies in voices (a kind of auditory "uncanny valley" effect). A classical example used in the audio coding community is the intro from Suzanne Vega's "Tom's Diner".

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.