Hi there,

So, I'm on a schedule to "finish" RTK by the end of May or a little sooner, so I'm about to enter the sentence phase. I know that having audio for every sentence is not realistic, however, I really really want to have audio for as many cards as I possibly can. I plan on having some friends record some sentences I've picked out sometime later on this summer, but until then I'm mainly gonna be grabbing audio from Smart.fm and anime/news/dramas.

The difficulty is that I don't know exactly how to "grab" it. With my old PC, I could just use Audacity's "What You Hear" function, but the sound card in my new PC doesn't support it. Most of my anime and shows are in MKV or OGM format. I would rather not have to rip the audio stream from all of them and then trim out the parts, but if that's an option, I'll take it. What I'd really like to be able to do is what I used to do with Audacity: just hit record, then play the line in the show, then hit stop on both VLC and audacity and voila, I've got the line.

How do you guys do it? Any tips?

Thanks!

If you use Linux you could route the signal with pulseaudio to an recording input unless I am very much mistaken. Jack should work, too.

thorstenu Wrote:If you use Linux you could route the signal with pulseaudio to an recording input unless I am very much mistaken. Jack should work, too.

I do! I'm running Jaunty now. How would I go about doing that? I'm a bit of a linux newcomer, so even Ubuntu is a bit foreign to me still. That's great though. Would I still be able to hear the audio, so I'd know which parts to record and everything?

sethg Wrote:I do! I'm running Jaunty now. How would I go about doing that? I'm a bit of a linux newcomer, so even Ubuntu is a bit foreign to me still.

Pulseaudio is some sort of linux newcomer, too. Its about audiostreams you can route like you want.

sethg Wrote:Would I still be able to hear the audio, so I'd know which parts to record and everything?

Everything is possible with pulseaudio. I tested it and it seems Pulsaudio makes an monitor stream the input section of the "Volume Control"(this app: http://0pointer.de/lennart/projects/pavu...enshot.png). You have to change recording in Audacity to alsa:pulse and then you can route (arrows pointing down in the screenshot) the monitor stream into audacity (Recording section in "Volume Control") -Audacity has to do some recording before you can see any option here.

EDIT: I hope the availability of the monitor stream doesn't depend on the soundcard.

iKnow obviously provides its own audio (KO2001 as well, consider that as a resource). For anime/dramas/news consider subs2srs. With the software that has come out of the fansubbing comunity it is probably easier to make a subtitle file for the sentences you need rather than to deal with capturing the audio yourself. You'll get images/video for free in the process as well. For manga/novels/newspaper, run it through a TTS (Misaki voice). Better than nothing.

I'm not sure if it is too late but on some sound cards, the "What you hear" feature is called "Stereo Mix". It works the same way. Sorry if you already knew that.

If you have a Mac http://www.rogueamoeba.com/audiohijackpro/ is awesome. The trial version won't capture more than 10 mintues at a time, but that doesn't bother us.

thermal Wrote:If you have a Mac http://www.rogueamoeba.com/audiohijackpro/ is awesome. The trial version won't capture more than 10 mintues at a time, but that doesn't bother us.

That is absolutely awesome. Thanks for that.