The closest I could think of is "low quality" but that is not as specific as I'd like, it could also mean that the music is bad.


I like this song, but you should really get the CD. This sounds _______

  • 3
    Please don't answer the question in the comments. Comments are for requests to clarify the post. – Matt E. Эллен Nov 20 '15 at 15:46
  • This is an unclear question as we do not know just what the problem with the sound is! It could contain static from radio or scratches from vinyl or compression artifactes etc.. As it stands I should be closed as 'Unlcear'.. – TaW Nov 23 '15 at 13:08
  • @TaW I think the question is clear and should stand – buildsucceeded Nov 23 '15 at 17:09
  • Well, then you surely can answer my questions, yes? No, you can't because the question is totally unclear. We are supposed to guess what the recording sounds like and why.. All those answers are based on just that: Guesses. – TaW Nov 23 '15 at 18:10
  • I would be interested to know where this kind of questions comes from. I guess some kind of books preparing for examinations. But I can't see that such questions help understanding language or vocabulary. – rogermue Feb 1 '16 at 23:18

24 Answers 24


"Downsampled" is the literal equivalent for sound. In both cases -- pixellation and downsampling -- analog reality is overly quantitized, leading to perceptible artifacts.

In the particular example sentence, though, I might say "this sounds like a bad rip."

  • 21
    "Undersampled" would better describe the situation imo. – user1306322 Nov 19 '15 at 11:18
  • 20
    For what its worth, the sound of downsampling is different and distinct to the sound of an over-compressed MP3. Most of the problems people hear with digital music are caused by over aggressive lossy compression rather than a change of sample rate (most compression doesn't change the sample rate). – JPhi1618 Nov 19 '15 at 15:03
  • 5
    @JPhi1618 Bitcrushing refers to quantization in the amplitude. The rough equivalent for images is posterization or color reduction. Downsampling refers to quantization of the time domain, and is indeed the better equivalent to pixelation. – nitro2k01 Nov 20 '15 at 8:48
  • 4
    Pixelated implies there isn't enough information to display the picture looking nice at the current size - e.g. trying to scale up an icon will look pixelated. Nitpicking, but an icon isn't "overly quantized", it's "sampled appropriately for icon size output". Is there any audio equivalent of an icon? There is low sample rate audio in normal use in some situations - makes me think I'd call it telephone quality. – TessellatingHeckler Nov 20 '15 at 18:15
  • 3
    Undersampling and downsampling are vastly different. An undersampled signal will have all sorts of distortion from high-frequency components folding back down as lower frequencies. A downsampled signal will just be missing the high-frequency components. – R.. Nov 21 '15 at 7:32

Lo-fi, from low fidelity:

  • the production or reproduction of audio characterized by an unpolished or rough sound quality (MW)
  • (of sound reproduction) of or giving an impression of poor quality (Collins)

Also lo fi and low-fi (TFD). Wikipedia:

a type of sound recording which contains technical flaws that make the recording sound different compared with the live sound being recorded, such as distortion, hum, background noise, or limited frequency response.

In digital audio, the term "lo-fi" usually refers to an audio file with a lower bit rate or sampling rate, and thus a lower sound quality.

  • 6
    This could be ambiguous though; these days "lo-fi" is often used to describe a desired quality in a performance or recording. – chapka Nov 19 '15 at 2:04
  • @chapka True. This term refers to low-quality or low-quality sounding music. Pixelated is similarly vague; could be that the image is created with a low-megapixel camera, or that conversion to another format reduced its quality. Lossy is a great word and refers more exclusively to quality loss due to data encoding. – Nathaniel Nov 19 '15 at 2:08
  • 3
    If lo-fi is intended, then so is the 8bit pixel games surely? So they really are synonymous. – RemarkLima Nov 19 '15 at 9:59
  • @RemarkLima that's slightly different/more limited. 8-bit sprites are to images as chiptunes are to audio. – JAB Nov 20 '15 at 19:03


From The Free Dictionary (emphasis mine):

  1. (Electronics) electronics a. an undesired change in the shape of an electromagnetic wave or signal b. the result of such a change in waveform, esp a loss of clarity in radio reception or sound reproduction

Distortion can be caused by a variety of methods, one of which is downsampling (which is digitally analogous to pixelation). Overuse of lossy compression can also cause distortion. It covers all use cases of "bad audio" that is not related to the actual music being played.

  • 2
    Even though this word is used in other contexts (overdrive/clipping), it is the technically correct umbrella term. +1 – Shaggi Nov 19 '15 at 13:35
  • Distortion in sound is nonlinear amplification, typically resulting in clipping and harmonics. Its equivalent in imaging would be overexposure and subsequent loss of contrast. – Gin Gordon Nov 21 '15 at 18:24
  • @gingordon Can you cite a source for that? Nonlinear amplification can cause distortion, but it is not the sole cause. – March Ho Nov 21 '15 at 22:35
  • @MarchHo Reference: Distortion in music -- Wikipedia More comprehensive definitions: The Audio Dictionary and Distortion, Electronic signals -- Wikipedia – Gin Gordon Nov 22 '15 at 10:16

Another (slightly more technical) option is "lossy". Reference

  • 2
    +1; this is perfect for compression-related quality loss. – Nathaniel Nov 19 '15 at 2:09
  • 7
    Lossy compression algorithms are often not noticed (MP3) meaning that the term isn't exclusive in terms of perceived quality but yeah, it's step in the right direction (if not the destination :) – Lamar Latrell Nov 19 '15 at 2:13
  • 5
    Not sure why this has so many upvotes, it's technically incorrect. Whether or not the compression used is lossy has very little to do with perceived quality. A lossy high-res image can look nice, while a lossless low-res image will be pixelated. – BlueRaja - Danny Pflughoeft Nov 19 '15 at 16:32
  • 1
    @BlueRaja-DannyPflughoeft Mostly because music doesn't follow that image metaphor. Almost all music is the same "resolution" of 44.1K or 48K samples per second, so when you hear a stream or file that sounds bad its typically because of the lossy compression used. If all images were 1280x1024, the only difference in perceived quality would then be the lossy aspect of the compression. – JPhi1618 Nov 19 '15 at 21:19
  • 3
    @BlueRaja-DannyPflughoeft, That's wrong. Bit rate is only variable because of lossy compression. Sound files have a sample rate (resolution) and a sample size (16-bit, 8-bit, etc). For instance, a phone call would be 8000Hz (sample rate) of 8-bit audio (sample size). Music would be 44100Hz @ 16-bit (very common). Since audio has a bit depth of its own (sample size), the image analog would be bit depth. – JPhi1618 Nov 19 '15 at 21:42

At first I thought garbled, but that's more about distortion.

Low quality sound, especially music, sounds tinny. An example is music on a telephone; the audio has been through a low-pass filter so it doesn't sound as rich as it should.


You may be listening to a sound that has been bitcrushed. This is an effect that can make something sound like it has been produced with an 8-bit or 16-bit soundcard. Listen to the sounds from this video for an example of what I'm talking about.

  • +1 from me and I considered this also (I know the plugins you're referring to). But to a layperson it's a step too far in jargon? – Lamar Latrell Nov 19 '15 at 19:32
  • 1
    What's wrong with a 16bit soundcard? That's plenty more depth than you need to cover the dynamic range of human hearing. xiph.org/~xiphmont/demo/neil-young.html explains why higher sample rates and bit depths are unnecessary. Sure higher bit depth is slightly better for studio use, but you can capture in 16bit and have no real problems. 8bit samples are very noticeable, though. – Peter Cordes Nov 21 '15 at 11:12


This is the generic term, applicable to any digital medium when describing some artefacts of low resolution.

More technically, aliasing is the stepping between one digital sample and its neighbour. In imaging, it manifests as stepping from a pixel of one colour to the next pixel ("pixellated"); in audio, it manifests as stepping from one displacement to the next.

Source: any good dictionary

  • 3
    It describes an artefact, but only one artefact - not an umbrella descriptor of all of them. – Lamar Latrell Nov 19 '15 at 19:32
  • 1
    but aliasing is one artefact (or 'cause' of an artefact). It's not a generic term. – Lamar Latrell Nov 20 '15 at 0:17
  • 1
    Arguably, it is possible to have downsampled audio (i.e. pixelation) without aliasing if a low-pass filter is applied prior to the downsampling. This is also the case in images – March Ho Nov 20 '15 at 0:32
  • @MarchHo, yip, aliasing only occurs if the sampled signal has frequency (temporal *and*/or spatial) components that are higher than half the sampling frequency/pixel dimension – Lamar Latrell Nov 20 '15 at 1:48
  • Indeed, only very bad audio processing will introduce aliasing. What's more likely is that the high frequencies have just been lost. – R.. Nov 21 '15 at 7:34


  1. simple past tense and past participle of overcompress


Since data compression is the actual phenomenon that causes the "audio pixelation", this is what you should refer to.

Data Compression

  1. (computing) Any of several techniques for reducing the number of bits needed to represent an item of digital data, either to save storage or to use less bandwidth when transmitting it


  • 10
    Note that compression is also a sound technique, so if you're talking with sound engineers, they will think you mean something very different. – DJMcMayhem Nov 19 '15 at 15:40
  • 2
    @DJMcMayhem good point! Perhaps, while referring to this, say "excessive data compression" or some variant of that. – Ben Leggiero Nov 20 '15 at 16:35
  • 1
    outstanding, a great answer. Regarding DJ's point -- you know, for better or worse homonyms are totally and completely normal in English, and that includes within technical fields. In contrast to the second half of your sentence, DJ, ANY sound engineer younger than say 95 years :) when asked what does "compressed" mean would instantly explain it can mean either data compression, or the old-fashioned meaning. Note that there is only in existence now 'compressed' audio, there is no alternative (other than some novelty analog recording on vntage equipment). – Fattie Dec 18 '15 at 16:23
  • 2
    Hi @Ben your answer is right, but your comment is wrong. Do not say "excessive data compression". Just say "overcompressed". Homonyms are utterly commonplace, including in technical fields. The word "overcompressed" (as used here) is an utterly commonplace word amongst audio engineers, you hear it every two minutes. It's like a car discussion including "horsepower". You do not need to qualify it, in any way, as it happens to be a homonym. – Fattie Dec 18 '15 at 16:41

I humbly submit warbled. I can't find a dictionary definition, but as the Google search shows it is definitely in use as the kind of thing you're looking for, and to me personally it refers to the kind of distortion you get on a bad GSM connection.

  • 2
    Not technically 'sound', but a good word for this (IMHO) – Lamar Latrell Nov 19 '15 at 19:34
  • 2
    Warbled is actually a specific kind of distortion only produced by records and tapes. I wouldn't use it unless I meant that specific sound. – Jasmine Nov 19 '15 at 23:10
  • 1
    @Jasmine: You also get a similar kind of distortion from very low bitrate with some codecs like AAC, and to some extent MP3. Another way to describe that is swimmy. – Peter Cordes Nov 21 '15 at 11:15
  • 1
    Swimmy/watery is different. Warbled is like the sound of an EU ambulance siren. I understand if you've never heard a warped record being played, this term might be applied to similar situations - common usage differs than technical usage. – Jasmine Nov 23 '15 at 22:24
  • @Jasmine Correct, common usage does not have to reflect the precise technical term. Jargon arises from common usage, and often the chosen word has multiple common meanings at the time it is abducted as jargon. Thereafter, the common meanings are not lost from the original word. Unless I'm reading it wrong, the question is about conversational usage, and not asking for help writing a technical paper. – Rich Dec 7 '15 at 19:57

Consider, crackly and muffled

crackly: having or making a series of short, sharp noises M-W

muffled: being or made softer or less loud or clear Vocabulary.com

  • 2
    Another similar word is fuzzy – Kidburla Nov 19 '15 at 12:17


Pro: Unwanted artifacts of digital storage and processing of music are often called glitches and are highly analogous to pixelation in practice: usually to be avoided, but occasionally used for effect.

Pro: The meaning of glitchy is quite intuitive, even when encountered for the first time.

Con: Not really analogous to pixelation from a technical standpoint.

  • Your answer would be improved by a dictionary reference. – Ben Nov 20 '15 at 15:45
  • The introduction of this wikipedia article en.wikipedia.org/wiki/Glitch_(music) uses the term in the context of music. A quick search of online dictionaries shows that they define the term more generally. I might be misjudging how widespread this understanding of the word is :) – gös Nov 22 '15 at 11:14

The first word I thought of was "grainy". This is often applied to pictures rather than sound, and some of the dictionaries I looked at only had the visual definition.

From the Oxford Dictionaries:

(Of sound, especially recorded music or a voice) having a rough or gravelly quality: the grainy sound of bootleg cassettes


ANSWER: Digitized
As "image is to pixalated" is an expression of an image in binary pixal (vs. point) and a song's equivalent of a binary digital representation of song is also 'digitized'

If pixelated is used as an adj. E.G.;'Man that image is totally pixelated, it looks like a bunch of blocks". In that case pixelated to image would be like distorted to song:
ANSWER: Distorted.

I believe the most commons sense iterpretation is the first answer though: DIGITIZED

  • I know that many dictionaries say that "splitting up an image into pixels to make it digital" is one definition of pixelated...however, having been a professional in graphical arts for well over a decade I have never once heard the term used that way in private or professional life. It is always used to refer to the artifacts (such as through overcompression/lossiness/scaling/etc.). The question, similarly, is referencing that usage and looking for the audio equivalent. So digitizing does not work here whatsoever. – Jimbo Jonny Nov 23 '15 at 15:52
  • a superb answer. – Fattie Dec 18 '15 at 16:24


Per Oxford Dictionaries, meanings 1.1 and 1.2:

  • (Of a voice or sound) rough; grating:
  • (Of a record) making a crackling or rough sound because of scratches on the surface:

The meaning presumably comes from the mechanical analog audio storage medium, but also carries forward to the CD (optical medium) reference given, e.g. due to scratching.

It may be less applicable for a digitally recorded file stored on another medium (e.g. a flash disk), as the distortion usually occurs during read from the moving medium's player.


An image becomes pixelated when the density of pixels becomes low enough for humans to see them. An audio becomes decimated when the speed of samples become low enough for humans to hear them.

"Decimated" is the most semantically correct term, IMHO. Though I've never heard an audio engineer or musician use the term.

I'd say "Down/Under-sampled is the next closest that would be used by a non-scientist. It's meaning is further away because it implies a comparison to another state. Under what? Where "Pixelated" and "Decimated" do not.

"Aliased" is more general, and able to refer to both pixelated and decimated. "Low-fidelity" is much more general referring to any number of defects.

Anything to do with compression or bit-crushing are good guesses. But they refer to reduction of precision in other dimension. Image analog to that is messing with the number colors that can be present in an image.

  • I wish the word under-sampled had more prominence in your answer because to me that is almost perfect. – Ubiquitous Nov 20 '15 at 11:17

For song, the closest equivalent to pixellated is vocoded, or for instrumental music it would be 8-bit. Though both of those would be even more apparent on the CD.

  • 2
    "Vocoded" often has a specific meaning that is very different from what the question is about. That is, "vocoded" would most often be used to mean "processed with a vocoder". – Todd Wilcox Nov 19 '15 at 13:51
  • Agreed - but it is a close match to the word "pixellated" in terms of the effect that the process has on the input. I'm not suggesting that it's the right answer for the OP, but it might be the right word for someone who's come here looking for an answer to the same question. – Simon Nov 19 '15 at 15:35
  • This isn't technically correct. Vocoding was originally (and still is as far as I'm aware) an analog technique. Source: I own two, and I was informed of their history via Alan Turing's biography. – Lamar Latrell Nov 19 '15 at 19:30
  • @LamarLatrell Meh, really? I dpn't think vocoding is an "analog" technique. Vocoding means modulating a carrier signal in a certain way ("blah blah frequency blah blah input blah blah"), how this is achieved - with analog filter banks or DSP - is unimportant in my view. Of course, vocoders were all the craze in the 70s and 80s. – Tobia Tesan Nov 19 '15 at 23:33
  • @TobiaTesan, I said 'originally' - and by that I meant to stress it isn't an artefact of digital compression, which in the context of this discussion is important (in my view). Digital vocoders, like those I've found in audio applications look like they are digital 'copies' of analog style encoders (band pass envelope followers modulating/amplifying a corresponding sawtooth/whatever) - i.e. an analog 'technique'. You're probably aware hey were originally devised to encode voices in a reduced parameterisation (more channels, but they had reduced temporal complexity).. heh heh, blah blah – Lamar Latrell Nov 20 '15 at 0:12

I would say that "pixelated" refers to an image that is "low resolution", which could be for a variety of reasons. So comparing an image to a song using "pixelated" for the image would be looking for a property of a song (music or generically "audio") that describes its "resolution". While such a property probably exists (I am not an audio engineer or even a hobbyist), it's probably not the property the OP really means to describe.

"Low quality" works for an image, and even works well when describing "audio", but as @Shel points out, it is not specific enough to describe the music or the song because a "low quality" song would most certainly be thought to describe the song in a different way than intended.

But taken in the context of the statement in the question:

I like this song, but you should really get the CD. This sounds (like a) "low quality" (recording).

almost works because the statement is clearly a recommendation of the quality of the song itself.

Some alternate wordings could be:

I like this song, but you should really get the CD. The sound of this (recording) is "low quality".
I like this song, but you should really get the CD. This recording is "low quality".

In a different context, the simple statement "This sounds low quality", may not work:

Oh, this song, I don't like it, it sounds low quality.

"Rough", "scratchy", "distorted" all work seem to work pretty well in most cases, but again, not related to "pixelated".


I was thinking "noted" with the song divided to each individual note.

  • 1
    I don't believe noted means that; it certainly will not be understood as such. – TimLymington Nov 24 '15 at 11:57

I'd probably say it's exhibiting encoding artifacts, or if you want it in a word - artifactful or encoding-artifactful. @HotLicks suggests artifactuous

You won't find that in a dictionary though...

  • Wouldn't that be "artifactual"? – Hot Licks Nov 20 '15 at 1:19
  • 1
    I was just thinking of "artifacty" or "artifacted", so I'll leave this as a comment instead of a separate answer. – Peter Cordes Nov 21 '15 at 11:18
  • @HotLicks: No, because it's not that the song is an artifact, or has the nature of an artifact, it's that it has artifacts. Thus an image is not pixely. – einpoklum Feb 1 '16 at 9:19
  • @einpoklum - So it should be "artifactous"? – Hot Licks Feb 1 '16 at 21:08
  • Ummm... what can I say, I can't really fault that, but I like my suggestion better :-\ – einpoklum Feb 1 '16 at 21:52

A distinctive series of sounds written using musical notation.


The correct answer is, of course, simply


or perhaps "poorly compressed"

If you have to compress a track to send to a friend, and it turns out crappy, ie if you compressed it too much, there is one - and only one - phrase you use to describe that, you say

"we compressed it too much, sounds crap"

Occasionally the word "overcompressed" might be used. 90% of the time a simple phrase like "compressed to much" or "made too small" is used.

  • Hi Nathaniel - if so, sorry I missed it in the megaclutter. (Even though I text searched.) Since the whole page is so nonsensical, it's probably worth repeating. I can bear the shame of downvotes, i think :) Merry Xmas! – Fattie Dec 18 '15 at 16:38
  • @JoeBlow +1 for the 'crap' suggestion. Have you considered discussing this particular question in Meta by creating a chat room? They'd be a good place to raise this issue. It's easy too, I myself created one today to experiment for one of my questions which was put on hold. – Jony Agarwal Dec 18 '15 at 17:07

If we were to copy the etymology of pixel I suppose a sound could be "sonnelated". I generally say it's either "low-fi", or just plain "f**ked up".

  • I wish I could downvote. Twice. I bet you can't justify the construction of "sonnellated". – Tobia Tesan Nov 19 '15 at 19:08
  • @TobiaTesan, I immediately thought 'oh pooh pooh!' when I read your comment, but I'm finding it hard to do what you suggest :). I could have sworn I've read this term (or something similar) in a DSP book, but for the life of me and google I cant find reference to it... – Lamar Latrell Nov 19 '15 at 19:39
  • @LamarLatrell: I'm usually a bit of a dick on SO, but it's nothing personal, I just think this is a terrible answer. Let me be clearer. OP asked for a single English word. Now, "pixelated" is not a pretty word, but it is a word. Like motorcade. It's disgusting, who coined it was trying to ape "cavalcade", in which - in his/her head - "caval" meant "horse" and "cade" meant... uh, something. But it is an English word. Here, you made up your own, "sonnelated". That's not an English word (0 Google results), hence the first downvote, and can't be justified by construction, hence the second. – Tobia Tesan Nov 19 '15 at 20:42
  • @LamarLatrell I can't find any occurrence of "sonnellated" in that SO thread :-| – Tobia Tesan Nov 19 '15 at 20:42

Adding this answer for completeness:


Source and SO moderated discussion here:


However, it doesn't fit with the example sentence, as CD's are sampled themselves...

Also, 'sampled' can refer to sampling in the spatial domain, i.e. pixels themselves.

I maintain that the (+17 votes) question itself is ambiguous.

  • 1
    CD sample rate and bit depth is high enough quality to be completely transparent to the human auditory system, though (if done well). See xiph.org/~xiphmont/demo/neil-young.html. There's a link to a really nice video covering some basics of digital sampling, with a signal generator and oscilloscope. (Yes, that's the same Monty behind Ogg Vorbis and so on.) Anyway, point is, sampling well enough that there's no perceptible quality loss has been doable for a LONG time. Audio quality problems don't come just from the fact that it was sampled. – Peter Cordes Nov 21 '15 at 11:24
  • @PeterCordes, I'm aware... I've been commenting on this myself above. I added the answer for completeness, I figured if we were getting it wrong we may as well get it wrong correctly :) – Lamar Latrell Nov 21 '15 at 19:56

Downsampled or Bit crushed

The difficulty when comparing the qualities of an image and a song is that they do not equate on a fundamental level. Therefore this comes down to interpretation.

A simple black and white image is a 3-dimensional signal. These three dimensions are x-position, y-position and brightness at this coordinate.

One way of representing a song, is by making an audio recording. A mono audio recording is a 2-dimensional signal. Where the dimensions are time and amplitude.

Pixelation of an image (or video) is a result of quantising the x and y position values. A piece of audio does not have these two traits, but the amplitude and time values can be quantised. Quantising the amplitude is known as amplitude resolution reduction or bit depth reduction in the digital audio domain. Quantising the the sample rate is known as decimation or sample rate reduction. Quantising of both at the same time is often referred to as bit crushing or downsampling a signal.

This video from an installation explores the analogy of video pixelation to audio bit crushing: http://tmblr.co/Z-a4pt1CeJ8B7 (most prominent after 2:30).

  • 1
    Welcome to ELU. Downsampled has already been given as the highest-voted and accepted answer. Please don't duplicate answers; it merely looks as though you're trying to cash in on someone else's good post [even if that's not the case. I doubt you did it deliberately]. You might usefully focus on bit-crushed though. – Andrew Leach Nov 22 '15 at 14:36

protected by tchrist Nov 24 '15 at 10:26

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

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