Tako, Reg, Kevin et al,
> Is the chap that was research assistant to Ingo Titze still on the list ?
Hmmm... yes, I am. Have been rather too busy to contribute for the last while.. and it doesn't look as though things are getting any less busy.
Tako's vocal clip..... >At the beginning and end of the clip, I am singing in "chest". In the >middle of the clip, I am singing in "head". Notice that when I am singing >in "chest" in the 200 Hz range, the formant range (for me, 2800-4000Hz) is >heavily populated by the 13-20th partials.
What I hear in this clip, and see in the spectrogram are two things:
1) A signal which is not good enough for analysis (signal to noise ratio high, recorded in an echoey environment or with reverb added, and some clpping taking place at some stage in the process). To get quality answers, you need quality equipment. The noise in the background looks like a combination of general equipment noise, and left-overs from the echo. Nothing more. It could also be a compression effect if you are compressing the audio at all (like recording on consumer MiniDisc). I don't think that from this recording anything can be said about the vocal tract non-linear noise generation ideas... Most healthy voices do not have much noise in them at all (ie Harmonics to noise levels of 16-20 dB for classical singing in mid-range, and up to 22 dB at high pitches). The noise components generally seem to come from the vocal folds themselves - when we see worse figures in the clinic, they can generally be attributed to thickened mucosal layers of the larynx (ie pathology or swelling). We use Harmonics to noise measurement as a routine quick vocal health check for singers particularly. Using our analysis package (which is not very reliable on such short segments, I get a Normalised Noise Energy of -6dB. That's the recording.
2) A normal countertenor arpeggio, shifting from chest into a rich head-type resonance. The shift in resonance is accompanied by an increase in the level of sound, which alters the visibility of the higher harmonics (particularly in the 4 middle notes in the arpeggio).
3) A relatively low level of "singer's formant" energy (measured as the ratio of energy above 2500 to the energy below). From the recording, the ratio is about 29% - about 10 percent below what we expect for classically trained singers. This is going to be the case with a ligter countertenor production, and explains why there is not much energy there in the spectrogram. With the quality of the recording, however, none of these figures can be overly trusted. We generally do our analysis recordings using Neumann microphones, outboard pre-amplifiers, and ensuring that background noise is at a minimum.
I prefer to think of the "singer's formant" as being a combination of things. In fact, I would challenge anyone to state that it is NOT due to a combination of things:
-> good, fast vocal fold closure will increase the energy in the higher harmonics. The sharper the closure and the louder the sound, the more energy there will be up in the higher partial regions.
-> appropriate vowel tuning will maximise energy transfer through the vocal tract. Without correct tuning for each vowel, you are fighting a losing impedance battle (especially going though the second passagio)
-> extra modifications (especially of the epi-laryngeal spaces) will favour the brightness created at the vocal folds, and help to develop that ring.
-> careful balancing of the tongue position to avoid LOW+BACK will help the ring to resonate, and also simultaneously give warmth to the sound.
Don't have time for any more comment! Sorry.
Martyn Clark Edinburgh Voice Centre |
|
|
| |