On Mon, 10 Apr 2006 16:18, you wrote: > Thanks to all who replied, but apart from running Scribe on Wine it looks > like a lost cause. > > Sorry Craig - I can't provide a sample because we don't have recording > software, only Olympus playback software for Windows. > > The Olympus foot pedal connects to usb or a com port. Nick I'm sure > you're right that the pedal just emulates keys, but the hands off control > is really essential and I don't think hot keys would work very well.
Actually, on the nch website, as well as the aforementioned 'scribe' is also a program called 'switch'. It is for converting audio to different formats. My wife does this sort of work too, and frankly it was easy to install the Olympus software on a cheap laptop which was running Win 98. Aside from that, I understand the problems clearly, and would be glad to assist creating a solution. As I see it there are two problems. One of these is created by Olympus, and the other is a general problem. Problem 1: Audio input. Despite the fact there are cheap MP3 thumbdrives with FM and recording facilities, a lot of places have dedicated dictation hardware. The Olympus stuff is pretty good, so you will encounter it everywhere if you are dealing with this sort of thing. As has been pointed out, the sound file format 'dss' is proprietary. Olympus provide a utility for converting DSS to WAV (Windows-only), and you can of course convert WAV to MP3 trivially if you want. Alternatively the 'switch' software from NCH is supposed to work under Wine, so that might be a good solution to Problem 1. Problem 2: The Transcription Environment Transcribers use a foot pedal to control the playback of audio whilst they are typing. Typical features are pause/play, fastforward and rewind. Having a foot pedal means that the hands can stay on the keyboard. Pressing special keys to control the audio is tedious, and reaching for the mouse is right out. I can think of many solutions to this problem, assuming the audio is available in a playable format (such as wav or mp3), and a controllable audio player is available (such as xmms, or mpg123). What you need is to get the hardware to control the software. The Olympus pedal connects via USB or serial. I'd suggest it would be trivial to make a small script that polls the serial port and sends commands to the audio player. No, I have not done it with an Olympus pedal, but I have done it with other similar hardware. If you didn't have a pedal, I can think of three ways to build one: Get USB keyboard. Hack keyswitches and wire in footswitches. Pressing footswitches generates keypresses. Get USB or serial mouse. Hack buttons on mouse as above. Hack existing keyboard. Wire foot switches into keys that are defined as 'hotkeys' for keyboard operation of transcription software. Foot switch input is then identical to pressing a hotkey. Oh, building a foot pedal is also trivial- get one from Jaycar, get an old sewing machine from the SuperShed, make something with MDF and microswitches (or even some hinged blocks of wood that press on mouse buttons on the floor). Finally, you need some shell scripts or a bit of a GUI to tidy all the background stuff away so that the user can use it without having to know what's going on behind the scenes. So, whilst I can see a few solutions, this is why I installed it on Windows. A
