MP3 Recorders?

ilbrec asks: "A friend of mine is a field biologist who studies frog calls. Recently when I was talking with him, he was interested in a MP3 recorder for his research use so that he can record the frog calls and analyze the sound on his computer later on. No, he does not care for DAT at all. He has one himself that he has so much trouble with. He is looking for a small device that takes at least 64 MB or so of solid-state removable memory that is easy to use. Sound recording quality does not have to be great, as frog calls can easily be analyzed at 96 kbps, according to him. No, he does not care for DAT at all. Does anybody know a good one?" Well, we have embedded devices that can playback MP3s...why not ones that can record live audio and use CompactFlash (or Smartmedia) cards, instead of the old cassette?

Comments

Why not MiniDisc?

[Wakko Warner]

If he can't stand DAT (he must not treat his deck well...), Minidisc is the next best thing, as far as sound quality goes. I wouldn't trust the DAC/ADC on an MP3 recorder, and I doubt the sound would come close to MD (let alone DAT).

- A.P.

--
* CmdrTaco is an idiot.

Re:MP3 compression is too slow

[whydna]

Yeah but do you remember the battery life of portable CD players from yesteryear. I remember my brother had one that took 4 AA batteries and still only lasted a few hours. Now you can pop in 2 (probably less) and get tons of playtime....

Even though the DSP may have been there it would have made the player too power-hungry.

-Andy

Why MP3?

[yabHuj]

Why does he want to have an MP3 gadget to record the sounds? I'd recommend a plain old music cassette recorder ("Walkman"-type, e.g. http://www.prodcat.panasonic.com/shop/product.asp? sku=RQ-L30&CategoryID=223). You have 45 or 60 minutes recording for each side, media is EXTREMELY cheap (compared to any digital portable). Plus the recorder itself is cheap enough (~ $40) to be replaced easily in case is is dropped accidentally into the swamp. Some of these things even have voice-activated recording, so you can set it near the frogs, go away and just wait. Transfer to PC is done with any cheap soundcard.

Lossy compression for scientific research?

[raygundan]

I don't know a lot about the project these sounds are being recorded for, but I would imagine that discarding huge chunks of audio information (how do you think mp3's get so small?) that human beings can't hear is probably not a good way to go. If he is doing computer analysis of the sounds, wouldn't it be better to analyze a non-lossy recording? That way, even audio cues that a person can't hear (and that would be discarded by a lossy psychoacoustic model tuned for human beings) could be analyzed. Who knows? Maybe frogs do weird things with phase or overlapping near-frequency tones that we simply don't hear.

It's the conspiracy

[dubl-u]

A recent Slashdot article pointed to a long essay by John Gilmore, originally found on cryptome.org where, among other things, he basically claims that MP3 players don't record because their manufacturers are afraid that they'll get the pants sued off them.

It seems plausible, but does anybody have facts to back it up?

Re:bottleneck is probably the compression

[jovlinger]

You can do wonders with an IC. Especially repetetive things like mp3 encoding can be sped up trivially in hardware by just pipelining the encoder (or even just replicating the encoding circuits and buffer).

This is exactly the sort of task a DSP or IC kicks a general CPU's ass at.

Re:bottleneck is probably the compression

[lizrd]

Are you getting realtime compression though? Anybody with a computer can encode .mp3s, it's just a matter of being patient. I've done it on my old P120 machine back in the day, granted a 5 minute track took 6 hours to encode and once I'd done that it took 100% of my processor to play it back, but it was cool. Now I've got an Athlon 800 machine and even with that kind of power it's just a little bit faster than real time in doing the compression.

I'd say that the biggest problem with doing direct to mp3 recording is going to be getting the power to do real time compressions and getting it small and low powered. For now I'd stick with minidisks, they're cheap, small and fast.
_____________

MP3 compression is too slow

[cperciva]

What your friend is asking for is a processor using under 500mW which can perform realtime MP3 compression. That simply isn't going to happen -- MP3 is simply too complicated to perform in realtime without using a lot of transistors, ie. lots of power.

That's why CDs use raw audio, incidentally: audio compression was well understood at the time, but needed more computational power than was available in audio systems at the time.

Re:MP3 compression is too slow

[Smitty825]

I don't think that speed would be that much of a problem. Texas Instruments has had DSPs that were capable of 2Giga-flops performance in 1995. (Incedently, those are still used in the DirecTV sets that people use today)

bottleneck is probably the compression

[StandardDeviant]

It seems on the face of it that the thing keeping an embedded device from recording straight to mp3 would be the cpu required to do so. I'm basing this on the amount of time it takes my p2-450 equivalent machine to encode from .wav (uncompressed sound, basically what you're friend would be getting) to .mp3. Of course, yes, I'm going to 128+kbps, but still, there is only so much cpu you can cram into a portable the size of a pack of cigs.

Naturally take my suspicious as to the unfeasibility of this with a grain of salt as I'm just a sysadmin/desktop/server-type person, not an embedded hardware designer (maybe there is some uber-l33t way of using a custom chip to do this). I'd have to cast my vote for minidisc too, it sounds quite nice (a friend of mine uses his portable MD recorder/player to sample stuff for his electronica band).

--
Fuck Censorship.

Re:bottleneck is probably the compression

[StandardDeviant]

No, I'd guestimate that it takes about 6-8 minutes to encode a 5 minute track on my machine. So I'm not quite to the real-time level yet. Partially I think this is because the FPU on my proc (a celeron 300a that's been running @ 464 for about 18 months now) isn't as good as a normal p2 and it's cache isn't as big. :-) But it was cheeeep, and that's real good if you're a po' college student.

My point was that it would take a pretty powerful CPU to encode to mp3 in real time. Namely a cpu you aren't likely to find in an embedded device due to cost, power, thermal, and space constraints...

yes, anyone can encode to mp3 given patience, but the problem is that data collection has to be in real-time (i.e. you can't tell the frog to shut up for a sec while your recorder chugs away...).

--
Fuck Censorship.

MP3 Recorders? How about Minidisc?

[vraptor]

I have seen some of the newer MP3 portables with the ability to record voice/sound/frog calls. But I don't think they are recording in MP3 mode--I think they are recording with no compression b/c the tech specs usually indicate they can only record a few minutes of voice.

How about Minidisc player/recorders? They have optical out...you could either get an optical->analog converter, or get a sound card which accepts optical link cables.

Re:MP3 Recorders? How about Minidisc?

[torinth]

How about Minidisc player/recorders? They have optical out©©©you could either get an optical->analog converter, or get a sound card which accepts optical link cables©

Yeah© I've got to go with a recommendation for minidisc recorders too:

1© The media's alot cheaper, ¥unless you have a laptop handy to download all the mp3's too©

2© They're about the same size as mp3 players / recorders© which is alot smaller than the dat recorders I've seen© ¥is size his issue? DAT is really good on other counts©©© :

3© I don't want to start a war here, but minidisc recordings ¥ATRAC-3 are higher quality than most ¥though, truely, not all mp3s©

4© Minidisc recorders with good solid state buffers ¥most have 40 second buffers, are as close to solid state as you probably need to get, if rough terrain is an issue©

5© And despite what 'vraptor' said, you don't need an optical link to get the sound out of the minidisc recorder© It's better, of course, if you can keep your D->A->D chaining minimal, but almost every minidisc player has an analog out line© ¥For headphones, if nothing else

Hope that helps!

-Andrew

mp3 is probably a bad idea

[wishus]

i would say any lossy compression scheme is probably a bad idea. (that includes mp3 and ATRAC).. These compression schemes discard frequences outside the range of human hearing, and quite a bit from inside that range (20Hz - 20kHz).

Of course, I am assuming he has a microphone with a frequency response > 20kHz in the first place.

I don't know why he hates DAT, but there are some CD-R decks out there now.. That's about it for him, though.

wishus
---

do slashdogs ALWAYS reply without reading links?

[Socializing Agent]

Expect to see products using this chip within 6-9 months.... As for products available at present, Idon't believe there are any, though the Nomad II will record voice using a built in mic at 32kbps.

Look, no one is saying that such a product is not feasible. However, the industry will keep such products from ever coming to market in the US or EU. Because of the SDMI failure, the music industry cannot control MP3, and thus will do everything to prevent it. Look at how they killed DAT if you need an example.

As for your assertion that the Nomad will record voice at 32 kbps, that is just plain ridiculous. You can get better fidelity than a 32k mp3 with a knitting needle and a wax record on a hot day. The question specifically states that the biologist needs at least 96kbps, and I'm willing to bet that the "built-in mic" is going to be horribly inadequate as well.

If there is a hi-fidelity MP3 recorder that allows ordinary consumers to produce non-watermarked mp3s and transfer them to computer within the next year, I will publically recant. However, there is no chance of this. Read the cryptome article if you don't believe me.

MP3 recorders will never exist

[Socializing Agent]

Look at this excellent rant for why mp3 recorders will never exist. Minidisc is your best bet. You can get up to 149 minutes of monaural on a single minidisc, and the results will be very high quality. I personally used a minidisc for ornithology research back at Uni, with great success.