10 Nov - Rishell & Raines at WUMB
It's rare for a Wednesday evening. Arlene has her printmaking class, but it was canceled for this week. There was a members' concert at WUMB radio, the UMass/Boston station, and she got tickets. The performers were Paul Rishell & Annie Raines, an unlikely looking country blues duo. He's just about 50, gray and balding. She looks about 30. They're terrific. He plays blues guitar and gets as many notes out of it as you'd ask of a piano. She plays mandolin sometimes, but mostly shows you that a harmonica is a real musical instrument. After hearing many folk musicians who hang a harmonica on a frame around their neck and play one riff on it between stanzas of a song as an afterthought, it was a revelation to hear what Annie Raines can do with a harmonica. They talked some about the blues musicians of the '20s and '30s whose songs they cover, but mostly played more music than you usually get at one of these WUMB concerts. This concert will air on the Sunday after Thanksgiving at 11 AM at 91.9 FM. We came home with two CDs, Moving to the Country and I Want You To Know, both on Tone-Cool records
Oh, and speaking of CDs, Glen had a copy of his new CD Oy, It's Good, at klez last night. He had been in Berlin over the weekend to play at the Berlin jazz festival and was pretty well jet lagged. They got to see some of the preparations for the celebration of the 10th anniversary of the Berlin Wall coming down, but left before the party proper started. In spite of the jet lag, Glen did say to our saxophonist, “I want you to play out. The saxophone is a loud, obnoxious instrument, so let's hear it.” I thought I was sounding good on trumpet for about the first four pieces last night. There's something I can only describe as relaxing and letting the sound come through that I'm able to do more and more often these days. When I succeed in doing that, the trumpet sounds like, well, like a trumpet is supposed to sound.
There was another mouse in a trap in the garage this morning, and two traps had had the bait stolen but weren't sprung. At least we haven't been hearing mice in the attic the way we did last fall. Since the cable modem cable runs through the attic, that's kind of important.
We had an interesting talk at work yesterday about pitch-invariant speeded-up playback of speech. If you record speech and play back the tape (or whatever it is) at a higher rate of speed, the pitch goes up and eventually you get Alvin and the Chipmunks. To play back at faster than real time, this guy's program was estimating the pitch period of the speech (that is, if the fundamental pitch was 200 Hz, the pitch period is 1/200 of a second) and dropping whole pitch periods, overlapping the remaining data and interpolating it to get smooth transitions. OK. Maybe I've left out too much here. We're talking about digital recordings, so you speak into a microphone attached to your computer's sound card. The sound card samples the microphone voltage 11,250 times a second (or maybe 22,500 times, take your pick) and turns the voltages into numbers, so there are lots of samples in 1/200 of a second. Let's see, 11,250 / 200 is 56 1/4. To play back at twice the speaking rate, you would play 56 samples, skip 56, play 56, and so on. Except not quite; you would take say the first ten of the 56 you're going to drop and mix them with the first 10 of the next 56, in a sliding scale. For the first one you could use 9/10 of the old sample and 1/10 of the new one, then 8/10 of old sample number 2 and 2/10 of new sample number 2, then 7/10 and 3/10, and so on, so you kind of fade out and fade in. Otherwise you get annoying clicks and pops where you splice different sets of 56. This is more arithmetic than I want to do, but a 330 Mhz computer can do a lot arithmetic in 1/11250 second; 330 million divided by 11 thousand is what, about 30,000 clock cycles? A lot of arithmetic. The nifty part was hearing the demonstration of how this worked with and without the smoothing (the fade in-fade out averaging I just described) and at different playback rates. It really sounded as though the same guy was talking faster or slower. There. I've been hearing all the technical details of electronic music from Ceej, and Mike Reed's synopsis of superstring theory; I just wanted to show you I can be as geeky as the next person.
 |
 |
 |
|
 | |||