trader's little helper ?

[Mills1988]

Is this pretty accurate to test a wav file ? to see if it's an mp3 or not? Seems pretty simple to use- just click on a wave file and press "test". No looking at spectral analysis (which I have no clue about). I've done a few files so far. It never says 100 percent that it's MPEG, but it did say 95 percent that the one file was MPEG. Sorry if this topic was covered before.

[procella]

TLH isn't accurate to test for lossyness. It detects some bad ones but it fails in others. If you really want to know there is no way around FA and SA.

[Five]

Originally Posted by procella

TLH isn't accurate to test for lossyness. It detects some bad ones but it fails in others. If you really want to know there is no way around FA and SA.

yes

[Mills1988]

Ok. Thanks. So it's not accurate to test if it IS MP3, but how about when it says it IS 100 percent lossless? There is a chance it still could be MP3 sourced?

[roann]

Originally Posted by Mills1988

Ok. Thanks. So it's not accurate to test if it IS MP3, but how about when it says it IS 100 percent lossless? There is a chance it still could be MP3 sourced?

Yes, there is. Unfortunately it is *impossible* to give a mathematical expression/formula that allows to determine for sure whether an audio source is lossy or not. This is due to the number of factors (and the links among each other) that the quality of an audio recording is based upon. If you'd know *all* these factors you could say not lossy with 100% accuracy. But you do not know all factors (for example SBD or FM or Pre-FM or TV or AUD, analog or digital recording, mics used, other recording equipment used, acoustics of the venue, tape generation, filtering or equalizing or remastering somewhere in lineage, and so on, and so on). As you can image all this is way too complex to be exactly captured by a formula or procedure. What the formula used finally gives you is a bit like the weather forecast ...

So Trader's Little Helper is a 100% accurate tool *except* when it comes to test a wav file for lossy source. And it's not TLH that's not accurate. TLH just acts as a graphical user interface for a free command-line application named auCDtect.

In my experiences it is reliable when it comes to studio recordings (btw that's what it was developed for), but it often fails on live recordings. Anyway, use TLH only as a part of the puzzle. As said in other posts above, if you really want to know there is no way around frequency and spectral analysis. But sometimes even such techniques don't give you an obvious result. Just read some threads here and you'll see that sometimes even the experts are arguing about how to interpret the results.