The process and equipment I use has evolved over time. I think I’ve found a method that gives pretty good results for me.
1). First, when I decide I like a solo enough to transcribe it, I download to my iPod, I burn it to a CD for the car, I edit the audio track to just isolate the solo section (if it’s a long track that has multiple choruses of piano, bass, etc.) and burn that to a CD, etc. The idea is that I listen to the solo over and over again before I ever sit down with paper and pencil. I try to get to the point that I can sing along with the solo. The analytical part of my brain starts to kick in as well and I can start to identify phrases without writing them down (i.e. ii-V outline, diminished scale run, major scale sequence, Bird quote, etc).
When I first started transcribing, I too often neglected this step. I would hear something, decide to transcribe it, and immediately head to the practice room and work it out phrase by phrase. I would struggle because I could get the individual phrases, but didn’t hear how they hung together as a whole. It also made learning them more difficult (see Step 6 below). I would occasionally remember bits and pieces, but would have a hard time playing through the whole transcription by memory. When you take the time upfront to really listen to the solo to the point where you can accurately sing along with it, you have internalized it to the extent that you grasp the whole thing.
2). After internalizing it by listening to it thoroughly, I sketch out the form. How many choruses, how many eights, fours, etc. I have a lead sheet with the changes.
3). I use Transcribe! for transcription (http://www.seventhstring.com/). At first, I was utterly dependent on the slowdown feature. I would slow down even simple phrases to 50%. Now having done so many, I can usually hear a lot of the things at full speed. I won’t BS - there are some double time runs that I have to slow down to finally get. But in general, my ears have gotten to the point that I try to work a full speed most of the time. What I like about this program is that you can loop sections over and over. The other nice thing is the EQ which you can adjust to boost up the solo or filter out other parts. This software makes life a lot easier - in the old days, I would use a mini-cassette recorder and constantly reverse the tape to work on a section. Working in a digital format with a looping feature is a major advance over the way I use to work.
4). I work with pencil and manuscript paper. I work with my horn. I know there are guys that can just transcribe with pencil and paper (or maybe with a piano for occasional reference), but I like to work with the horn in my lap. It helps with figuring out particularly “saxophone” type of things (false fingering, scoops, wide vibrato, etc)
5). Once I get it down with pencil, I transfer to Sibelius. I try to get pitches exactly right. For rhythms, I go for readability over precision. I won’t use 32nd notes, even if that’s what I really hear. I may put in an indication like “lay back” etc. to communicate the feel without stressing out over whether each 16th is exactly in the right place. I include the “vanilla” chord changes. What I mean is that even if the soloist is playing a line with b9, #9, #11, etc. I’ll just notate the chord as C7 (rather than list the extensions). Again, since these are intended mainly for my use, I stress readability over nailing every last detail. Also, as far as readability, they are consistently formatted to 4 bars per line (occasionally 2 bars if it is a long double-time run). I try to be good about following the rules of engraving as well (correct enharmonic accidentals, divisions of rhythms, etc.)
6). Finally, there is the learning process. I often will play through the solo at 50% speed and then at full speed. I try to play through the whole thing. I then isolate parts. I may work on 1 chorus per day. I will also isolate certain portions (e.g. ii-V’s) and practice them in 12 keys (see examples on the Online Lessons page).