Hello, everybody.

I'm desperately searching a digital voice recorder working under linux. I'm using linux professionnally in office (I managed to convince my partners to switch from windows to linux) and everything works (almost) perfectly except this problem.

I need a voice recording device that allows me to dictate and to transfer files to my secretary which can hear the file.

What I search is a device that:

- when possible is recognized as a usb mass storage device (but if another access method is easy supported, it's ok for me. I mean easy, 'cause my partners are able to click on a nice icon, but commands like "mount -t usbfs /dev/sdb /mnt/removable" are ways to complicated for them),
- record files in a recognizable format,
- allows me to edit what I've just dictate (like a word processor).

My tests until this day:

- for now, every Olympus device seems to use a proprietary file format (DSS) that can't be converted on linux (my goal is to play them in mplayer that I can command with customisable keyboards shortcuts);

- I tried some MP3 players. OK for mass storage device; OK for file format (mp3 or wma) but you can't do voice edition (that means "this is what I'm talking about ... er ... (rew) ... speaking about" without closing first mp3 file and creating a new one).

Has anybody an idea ?

Thanks in advance

Olivier

It's shame about the DSS formatting.. what are those Olympus fools thinking when there are so many good audio formats that are open standards?

I've seen many mp3 players with voice record features, so I don't know why you would have trouble finding one that just acts like a USB drive. Your mount command can be done away with for dumb users if you form the /etc/fstab file correctly.

The problem is sure not about usb-drive acting. Almost every mp3 player has this feature. The problem is about voice edition. Being able to record voice is one thing, editing what you have just dictated is another. With an analogic tape, if you make a mistake, you can just rewind and dictate the correct word over the erroneous one. With most mp3 player, every time you stop recording the device save a file, say 1.mp3, and if you rewind and dictate again, it will save a new file (say 2.mp3). If you don't have a correct text prepared in advance (and I never have) you will have plenty of files, which isn't practicable for the secretary.

Worse. What you've dictated and is erroneous won't be erased. Example: you want to dictate "this is a sentence that has to be dictated". You begin to dictate "this is a word that... ". You realize that you've made a mistake, rewind, and dictate again ".. a sentence that has to be dictated". What will you have with a conventionnal mp3 player ? Two files:

1.mp3: "this is a word that... "
2.mp3: ".. a sentence that has to be dictated".

and you can't just erase "a word that": you can just erase whole 1.mp3 or 2.mp3.

That's what I mean with voice edition.

But if you know any device, mp3 player or dictation system, that can do this kind of voice edition (and is usable under linux, which means standard format and usb drive), I REALLY would be gratefull.

By the way,

It's correct (and if I ain't too dumb myself, I won't use -t usbfs but -t vfat ;-)). If it's a standard USB drive, this will work automatically (say it's ok and pray). But if it isn't, the risk is that the drive is recognized once as /dev/sda1, another time as /dev/sdb1 and so on... (I had this problem with a Minolta Dimage E201). So I'll have to write say ten lines in /etc/fstab (one for /dev/sda1, one for /dev/sdb1, ...), which means that users have to test themselves which device name is used. And this is complicated for them...

Anyhow, thanks for having replied.

Olivier

The only options I can think of would be:

- Use a tape recorder ((micro)cassette, VHS, DVHS, or DAT), and when the product is finished, play it back to the sound card.

- Use the digital recorder, and pro tools or soundforge type of tools to do the editing digitally; in which case it will be easier if you just let the recorder roll, so you have one file to edit later.

- Maybe if you use a desktop CDRW recorder (like a musician would use), you might be able to backup up and start over from a cue point. In the end you'll have a redbook disc.

There's a way around that. I've heard that some folks are able to make mount points static based on the volume label of the drive that gets attached. So you could supposedly connect up a handfull of USB drives in any order, and they will always take the same mount points.

Specs of some (new?) Olympus digital voice recorders say they record in WAV format.
E.g. the W-10: http://www.olympusamerica.com/cpg_se...oduct=894&fl=4