What to do when the exercises get hard

Discuss EarMaster, ear training, theory or music in general, ask questions and share your experience.

Moderator: Quentin

Post Reply
User avatar
saxmaam
Rhythm ninja
Posts: 45
Joined: 12 Oct 2007, 10:41
Location: New Mexico, USA

What to do when the exercises get hard

Post by saxmaam »

Until recently, I didn't have a high opinion of the microphone as an input for pitches in EarMaster. I thought the recognition algorithm was defective, since I would have to sing LONG tones and the needle on the tuner would wag around wildly.

The problem was that I couldn't hold a steady tone with my voice.

Lately I've improved that, and I've realized that the microphone input for tones might be one of EarMaster's finest features. That's saying a lot, because I think EarMaster is a great piece of software.

What I am suggesting is that when the exercises get hard, stop being a passive listener and produce some tones yourself. If chord recognition is hard, sing those tones into the microphone. You may find that you aren't singing the right notes, which suggests that you aren't hearing them correctly.

The interval singing module is great, too, though there isn't a tutor for it.

Is anyone else getting mileage from singing the inputs?
User avatar
Arpeggio
Good listener
Posts: 9
Joined: 19 Jan 2011, 03:21
Location: UK

Post by Arpeggio »

I was slightly supristed when I tried to hold a note while watching my guitar tuner, although I get the note the dial is going up and down. I find higher notes easier to hold I guess because of all my untrained vocal cords those are smaller and have less movement with smaller vibrations.
User avatar
Quentin
EarMaster.com
Posts: 2339
Joined: 27 Apr 2007, 01:25
Location: Denmark
Contact:

Post by Quentin »

Higher pitches are easier to detect for tuners and software pitch detectors because the movements of the sound wave are much faster over the same period of time than those of a low pitch, which means that the calculation of the pitch average over a same timeframe can be made with much more data when higher pitches are involved. The pitch detection of higher pitches is therefore easier, faster, and more accurate than that of lower pitches, especially if the lower pitches are sung slightly inaccurately and/or with vibrato.
- Because in Music, We're All Ears... -
User avatar
saxmaam
Rhythm ninja
Posts: 45
Joined: 12 Oct 2007, 10:41
Location: New Mexico, USA

Post by saxmaam »

I'm thinking EM could do better at this. My Korg CA-30 tuner has NO problem with low pitches. SmartMusic certainly does a good job of pitch recognition on even very short notes.

I was singing some low (for me) notes for EM this morning -- C and B below middle C -- and the tuner was indeed jumping around a lot. I also tested with a no-vibrato synthesized tone. I'd think some simple windowing/filtering/averaging could improve the tuner's performance.
User avatar
Quentin
EarMaster.com
Posts: 2339
Joined: 27 Apr 2007, 01:25
Location: Denmark
Contact:

Post by Quentin »

Hi,

Have you tried to calibrate the microphone while singing low tones? There are several settings that might help to a very large extent (sensitivity, algorithm used, etc.). To calibrate your microphonne, simply go through the Mic Tuning Wizard from the TOOLS menu.
- Because in Music, We're All Ears... -
User avatar
saxmaam
Rhythm ninja
Posts: 45
Joined: 12 Oct 2007, 10:41
Location: New Mexico, USA

Post by saxmaam »

I'll give it a try and let you know what happens. It's been working for me as-is, but I believe I have to sing those low notes longer before EM accepts them.

Thanks!
User avatar
saxmaam
Rhythm ninja
Posts: 45
Joined: 12 Oct 2007, 10:41
Location: New Mexico, USA

Post by saxmaam »

Following up on the microphone issue.

For the microphone I usually use, both even and odd harmonics works the best. Low notes are jumpier. If I try just even harmonics, the tone is identified one octave higher than correct.

I took a different microphone on travel with me -- a condenser type I think, rather than the dynamic mic I usually use. It's a disaster. The two options with odd harmonics both lead to the tone jumping all over the place. The even harmonics option leads to identification one octave higher than the sung note. I wound up holding the microphone to my throat using even+odd to get correct identification.
Post Reply