30 September 2010

Audio questions

Frequent readers might have noticed that I am an obsessive podcast listener.

About a year or so ago Apple upgraded the software on my iPod to enable me to listen to podcasts and audiobooks at double and half playback speeds.   I really like this feature, since I can absorb twice the information in the same amount of time.  However, I have two questions.  The first is rather mundane and the second I find more intriguing more interesting.

Question One: Why can I only do this directly through the iPod and not when listening through iTunes?  Or am I missing this feature?  The button to enable this mode is very prominent on the iPod, but I have not seen one in iTunes.

Question Two:  I usually use double speed playback mode when listening to podcasts, especially interviews, since they are typically more slowly paced than programs which are pre-scripted.  When I play back at high speed while listening through headphones I have little trouble keeping up.  But when I play the same file back at the same speed through speakers, either on my desk or in my car, I have to really concentrate to be able to keep pace.  Why would there be a difference between headphones and speakers?



  1. To the second question, I have noticed a similar effect when listening to Spanish programming...or something similar. If I listen to mi telenovelas or Sabado Gigante REALLY LOUD I don't have trouble keeping up. But at normal volumes it's more difficult. I've always suspected that with reduced social cues (in my case, a cultural gap; in yours the higher speed mutes normal audible signals) a more intense sound signal allows me to concentrate harder. Basically I'm saying, reduced ambient noise helps you concentrate. In both cases I would suspect that were using high volume or headphones as a crutch and if we stopped, we'd get better at not needing them.

  2. Good theory.

    I'm going to (potentially) complicate this by saying that I often listen with only one of the headphones in my ear at a time. I haven't oticed a difference between both and just one, but now I'll be on the look out for that.

  3. The Simpsons. They speed up speaking routinely. Apparently the limiting factor isn't understanding, but producing understandable speech.

  4. Well speech intelligibility is highly influenced by the signal-to-noise ratio, and in the case of speech any sound that is not the speech is noise. So it may be the acoustics of your room that are causing the difficulty by adding in reflections that interfere, or there are other sources of noise like equipment fans and HVAC that are more easily distinguished from the program in Headphones.

    Also, we know that when people are listening on speakers and then put on headphones and turn them up to a subjectively equivalent level, they are actually creating much more SPL at the eardrum with headphones than with speakers, yet subjectively the level is the same -- so that might explain a SNR boost at the ear.

    Third, the harmonizing process necessary to fix the pitch of the double-speed speech is not without artifacts, and that means there will be missing cues in the speech as well as some other noises that would mask cues.

  5. Thanks for the explanation. That makes some sense.

    But if my earbuds provided such superior SNR, wouldn't purist audiophiles prefer those to actual speakers? I've been hearing for years about how shitty these sorts of headphones are and how you're not really hearing your music "right" if you use them.

  6. Some earbuds are fantastic. The problem is stereo recording.

    If you listen on loudspeakers from the "sweet spot" then a good recording will have the various instruments spread so that some are center, some are left, some are right. There is a whole range of "phantom" positions between the loudspeakers in front of you.
    That said, $600 Stax headphones will sound far better than any $600 set of loudspeakers, and I'm sure your headphones sound better than any speakers costing an equivalent amount to what you paid for them. Because the dimensions and power requirements of headphones are much smaller, it can be much easier to get very low distortion and other very admirable attributes.
    On headphones the stuff in the center (lead vocal, bass, kick-drum) will often sound like they are right in the center of your head, and the range of positions is from one ear to the other through your head from ear to ear. Most of us don't find that is pleasing or natural.
    In a good room with decent speakers and good recordings you can close your eyes and not know where the speakers are, they disappear sonically. The "in-head localization" effect of using headphones is harder to get rid of.
    The reason? Well, with headphones you don't have any "crosstalk" between the two channels and the two ears. With speakers you hear the signal from the left loudspeaker in your left ear and a little later the signal from the right speaker in your left ear. With tonal signals, the way the two waveforms combine at each ear converts the level difference between the to speaker channels into a signal phase difference between the two ears -- and that matches what some smart dude named Lord Rayliegh said 100 years ago about how we can tell the quacking of a duck is off to the right somewhere -- because of the slightly longer time it takes a sound to get to the farther ear if it comes from one side, our brain interprets the signal phase difference as a cue to where the source of the sound might be.

  7. Thanks for all the answers. I'm learning some good stuff here.