An Interesting Experiment

An Interesting Experiment

Joined: January 1st, 1970, 12:00 am

April 16th, 2011, 1:03 pm #1


On April 5, 1929, Bert Lown and His Loungers recorded the classic Tom Delanay tune "The Jazz Me Blues." The Loungers were Frank Cush, t; Miff Mole. tb; Rudolph Adler Johnny Costello, cl as; Paul Mason, cl ts; Chauncey Gray, p; Tommy Felline, bj; Ward Lay or Al Weber, bb; unknown, d. Listen to Lown's recording.

www.jazz-on-line.com/a/mp3c/HAR148180-3.mp3

It moves at a very good clip.

Of course, "The Jazz Me Blues" is strongly associated with Bix. It was Bix's first recording - Wolverine Orchestra, Feb 18, 1924. Bix recorded it again with his Gang - Oct 5, 1927. Bix's recordings were taken at a slower tempo than Lown's. 

One of trombonist Dymitr Markiewicz friends, slowed down (without changing key) the tempo of Lown's recording to match the tempo of Bix's recording. Here is the result, kindly sent by Dymitr.

bixography.com/JazzMeBluesLownSlowTempo.mp3

Frank Cush's sounds more Bixian in the slowed down version than in the version at the original tempo. And is it possible that  Miff Mole's solo sounds better at the slower tempo?

Thanks Dymitr.

Albert


Quote
Like
Share

Glenda Childress
Glenda Childress

April 16th, 2011, 2:11 pm #2

You're right. Actually, they <em>all</em> sound better!
Quote
Share

alex revell
alex revell

April 17th, 2011, 2:02 pm #3

On April 5, 1929, Bert Lown and His Loungers recorded the classic Tom Delanay tune "The Jazz Me Blues." The Loungers were Frank Cush, t; Miff Mole. tb; Rudolph Adler Johnny Costello, cl as; Paul Mason, cl ts; Chauncey Gray, p; Tommy Felline, bj; Ward Lay or Al Weber, bb; unknown, d. Listen to Lown's recording.

www.jazz-on-line.com/a/mp3c/HAR148180-3.mp3

It moves at a very good clip.

Of course, "The Jazz Me Blues" is strongly associated with Bix. It was Bix's first recording - Wolverine Orchestra, Feb 18, 1924. Bix recorded it again with his Gang - Oct 5, 1927. Bix's recordings were taken at a slower tempo than Lown's. 

One of trombonist Dymitr Markiewicz friends, slowed down (without changing key) the tempo of Lown's recording to match the tempo of Bix's recording. Here is the result, kindly sent by Dymitr.

bixography.com/JazzMeBluesLownSlowTempo.mp3

Frank Cush's sounds more Bixian in the slowed down version than in the version at the original tempo. And is it possible that  Miff Mole's solo sounds better at the slower tempo?

Thanks Dymitr.

Albert

What a fine record. Apart from the absense of Bix in some ways as enjoyable as the Wolverines version. I'm intrigued as to how it was slowed without altering the key. Technical wizardry I suspect.
Quote
Share

Joined: January 1st, 1970, 12:00 am

April 17th, 2011, 3:27 pm #4


http://audacity.freewarecentral.net/

A complete manual for using audacity is available in http://audacity.sourceforge.net/manual-1.2/

As to how precisely the software achieves changes in tempo without changing pitch is something I have not studied,  and I am sure I don't have the background knowledge to understand without learning a lot more of basic principles of acoustics, digital techniques and computer programming.

Albert
Quote
Like
Share


Brad Kay
Brad Kay

April 19th, 2011, 7:22 pm #6

Actually, Albert, with respect to fixing the Charles Dornberger "Tiger Rag" with "Audacity," two steps is one too many. Just slow it down until it's in B-flat. The tempo will take care of itself. This is all you'd need to do with the original 78, anyway!

I've used "Audacity" for the last several months to comb out clicks and pops from 78 surfaces, and now swear by it (when I'm not swearing AT it!).

The latest result is the revised WILLARD ROBISON disc. Took about a month, but now about a million ticks have been exterminated from the master, and it's a whole new experience. More to follow...

-Brad K
Quote
Share

Joined: January 1st, 1970, 12:00 am

April 19th, 2011, 9:27 pm #7


.... of Audacity. Of course, changing just the speed will change correspondingly the pitch, just like when we change  the speed in a turntable with variable speed capability. 

The change in speed needed to achieve the desired change in pitch (exactly one semitone) was calculated in my two-step procedure posting 

http://www.network54.com/Forum/27140/message/1297193479

using  the "Pitch Shift Calculator,"

<em>"The Pitch Shift Calculator converts speed and tone changes (RPM, IPS, Hz) to semitones and cents so that accurate pitch correction can be performed." </em>

http://www.theaudioarchive.com/TAA_Resources_Pitch.htm

In summary, the two-step procedure using Audacity in conjunction with the "Pitch Shift Calculator," although admittedly an over-kill, was very useful in that it illustrated clearly the concepts (tempo, pitch and speed), and how to manipulate them, implied in Vince's posting <em>"recorded slower-played back faster"</em> about Charles Dornberger's "Tiger Rag."

http://www.network54.com/Forum/27140/me ... 296690434/

Albert
Quote
Like
Share

Brad Kay
Brad Kay

April 20th, 2011, 12:27 am #8

Here's a better challenge: You recall my stereo comparisons of Bix solos, with simultaneous playback of alternate takes. At the time, my only recourse for synchronization was to change the speed of one of the takes, to match the tempo of the other. The resulting differences in pitch made some of those comparisons barely listenable and others impossible. Now with Audacity, that hurdle is yesterday's mashed potatoes. Any of them can be synched perfectly. It's just the matter of figuring out HOW.

Some of the Bix alternates with widely varying tempi:

the two takes of "Clorinda"

two of the three takes of "From Monday On"

"Loved One"

"Proud of a Baby Like You"

"Lazy Daddy"


best,

Brad K
Quote
Share

Joined: January 1st, 1970, 12:00 am

April 20th, 2011, 4:47 pm #9


Create wave files of the two solos. These have to be exact: from the first note to the last note of the solos. Measure the length of the two solos. Using Audacity adjust the length of one of the solos to match the length of the other.

Specifically. Assume the takes are T1 and T2 of lengths L1 and L2. Open file T1 in Audacity, select it, and using the "change tempo" (no change in pitch) utility, enter in the box "length (seconds)" the length L2 of Take T2. Save. Call this T1M (for T1 Modified). Merge T1M and T2. Now they should match in length.

What do you think? If this seems to be a useful procedure, I will try it with one of the recordings you sugggest.

Albert

Last edited by ahaim on April 20th, 2011, 5:37 pm, edited 1 time in total.
Quote
Like
Share

Joined: January 1st, 1970, 12:00 am

April 21st, 2011, 2:02 pm #10


In the process of syncing two music files, before merging, it is useful to import (click Project/Import Audio) the two files in audacity. Now you can listen to the two files simultaneoulsy. If they don't begin exactly at the same point, then a useful function is "Time Shift" (the icon is a horizontal double arrow). You can time-shift either of the two files so it coincides exactly with the other.

Albert
Quote
Like
Share