I just began using Audion 3 to convert some old tapes to MP3 (Lame 3.92), and discovered today that Apple's recently released "QuickTime 6" now does AAC encoding and playback. I'm wondering if the method I'm using with Audion-3 will also work for QT6:

In A-3, I run an RCA-to-minijack cable from the tape deck to the sound input jack of my Mac G4, and (after clicking the appropriate buttons) A-3 converts the tape analog signal into an MP3 file in one step. (without first making an AIFF file)
Then I use the MP3 Editor in A-3 to split the large MP3 file (with over 90 minutes of music) into smaller MP3 files, one for each song.

Will this two-step process also work with QT6, to produce separate AAC files for each song? And will everything be analogous to what I've been doing with A-3 and MP3?

I'm assuming it will work, based on information (in the Apple website) about QT6 - they say that it imports "sound" (like a stream of sampled tape signals, rather than a file?) and that you can edit files (I'm assuming the "sound file" can be independent, and it doesn't have to be connected with a movie), but if any of you has actually used this part of QT6, I'd appreciate knowing about it. Thanks, in advance, for any help you're able to provide.

Mike R

I'm not sure if QT can split MP4 files, haven't seen that feature yet.

Menno

You could do that easily in QT6 if you have the Pro licence, just select part of the track then do a cut and paste in a new movie and save as self contained movie...

Thanks. Yes, I meant Quick Time 6 Pro (which costs $30), not Quick Time 6 (no cost).
So an audio track (with no video information) must be saved as a MOVIE? Does this work OK? Does this make the sound-file bigger than it would otherwise be, to have it capable of also including video information?

another question:
Is the AAC encoder in QT6-Pro fast enough for real-time encoding of the stream coming in from a tape deck? I'll be encoding on a 400 MHz G4, on which Fraunhofer is very slow (the 96 kbps "normal" mode is the fastest I can use) and LAME-CBR is fast (320 kbps in the "best" mode is OK), but LAME-VBR in the "best" mode of Audion 3 (I'm not sure what the actual encoder-settings are) is too slow. When the encoding is too slow and it can't cope with the rate of the stream flow, the file contains "skips" (where the slow processing lost part of the fast data stream) and I have to use a setting that places a lower "bitstream and quality" demand on the computer.

also:
Does QT6-Pro give you a choice of sampling rates? I don't know much about this, but on a web-page I read that sampling at 32 KHz (instead of 44.1 KHz) would give better results for tape, since it would lessen the tape hiss. (many of my tapes are reel-to-reel without Dolby, a few are cassette without Dolby, and many are cassette with Dolby)

Mike

Thanks. Yes, I meant Quick Time 6 Pro (which costs $30), not Quick Time 6 (no cost).
So an audio track (with no video information) must be saved as a MOVIE? Does this work OK? Does this make the sound-file bigger than it would otherwise be, to have it capable of also including video information?

well, every file in QT6 is called a movie if it contains video, audio or both, so if you open an aiff file in qt and you do a cut and paste in a new file, this will be saved as an aiff file too.

another question:
Is the AAC encoder in QT6-Pro fast enough for real-time encoding of the stream coming in from a tape deck? I'll be encoding on a 400 MHz G4, on which Fraunhofer is very slow (the 96 kbps "normal" mode is the fastest I can use) and LAME-CBR is fast (320 kbps in the "best" mode is OK), but LAME-VBR in the "best" mode of Audion 3 (I'm not sure what the actual encoder-settings are) is too slow. When the encoding is too slow and it can't cope with the rate of the stream flow, the file contains "skips" (where the slow processing lost part of the fast data stream) and I have to use a setting that places a lower "bitstream and quality" demand on the computer.

I never tried that but I feel that QT isn't designed for that so I would suggest to do an aiff recording before...

also:
Does QT6-Pro give you a choice of sampling rates? I don't know much about this, but on a web-page I read that sampling at 32 KHz (instead of 44.1 KHz) would give better results for tape, since it would lessen the tape hiss. (many of my tapes are reel-to-reel without Dolby, a few are cassette without Dolby, and many are cassette with Dolby)

Not really, only possible selections are kbps/sec and mono or stereo but I have noticed that 44.1kHz starts only at 128 kbps, 96 and 112 is definitely at 32kHz. If you need more fine tuning for mp4 I would suggest you give MediaPipe's AAC encoder a try (this works only from file too)
http://mediapipe.sourceforge.net/

Frank

Mediapipe uses FAAC. Quality is VERY suboptimal. Just FYI...

Thanks for this info and it is even slower than QT too :-)

Frank

Audion's Fraunhofer encoder is awesome for sound quality if you put it in the highest quality (slowest) mode. I'd recommend just recording your line-in to an uncompressed 44.1khz 16-bit stereo AIFF / system sound / WAVE first and then using highest quality encoding at 128kbit or better in full stereo (not joint stereo). This will give you the least loss from the MP3 encode. I would also especially recommend this when you have a high quality source too, like CD ripping. The sound quality is definitely worth the wait (I have a 400MHz G3, so I know the encoding takes a while).

Fraunhofer? 128kb/s? Full stereo?
You`re still stuck in the middle ages

FhG MP3 encoder from Adobe Audition 1.0 VBR quality 40, "Current - Best" codec: http://www.rjamorim.com/test/mp3-128/results.html


Not very impressive

Raptor says, "I'd recommend just recording your line-in to an uncompressed 44.1khz 16-bit stereo AIFF / system sound / WAVE first and then using highest quality encoding ... The sound quality is definitely worth the wait."

Yes, this is what I've been doing. In late-2002 (you'll notice the original posts were August 2002, which was indeed the Middle Ages of recording/encoding for AAC) I stopped "on the fly encoding directly to mp3 or AAC/mp4" so I now record an AIFF file, then edit it with SparkME for Mac OSX -- unfortunately it's been discontinued ( http://www.tcelectronic.com/SparkLE ) -- and then (as you recommend) I encode it using whatever codec I want, even if it's slow. While it's encoding, I'm using my laptop to be productive, so I don't care about the speed.

Mike

adobe audition has an mp4 plugin does it not, where you can edit your mp4 files

That site clearly states they forced VBR on codecs not designed for it.

By the way, Audion 3 does not include the Fraunhofer encoder! It instead uses Coding Technologies, and the quality is back down there with the free MP3 encoders.

For audiophiles who don't want to keep their music in AIFF form, go with Audion 2 (I use Audion 2.6.1). And as I said before, quality Best, full stereo, 128kbps or better (I always use 192kbps to be sure it's as good as possible).

Even at 192kbps I noticed other MP3 codecs (Coding Technologies, iTunes, LAME) weren't sounding as clear in the treble as Fraunhofer. They sound good, but not as good. My test was actually listening to a few rips, not from a waveform analysis; and I believe what you hear counts most.

On the other hand, I'm on a dual 1.8 G5 now, and it still feels a little sluggish encoding in Fraunhofer. So you have to really want the best quality to put up with the wait, but it's definitely worth it to me.

This is a very bold and controversial statement, especially when it isn't backed up with ABX test results. As is recommending full stereo for 128kbps over joint stereo. Posts like this have no place on this forum.

Come on eofor, he did say that "what you hear counts most". It is only a question of building up on this statement and realise that an ABX test is the best tool to find what you really hear. And which encoder gives nearest to the original/least annoying results.