M2 Internship – Measure of vocal tract resonances during singing

The internship offer is part of the ANR-funded AVATARS project, aiming at studying voice production with a emphasis on larynx and vocal tract articulation. Voice production relies on three major functional stages:
1. at the thoracic level, air is expelled from the lungs
2. within the larynx, a flow-structure instability induce the onset of vocal folds, thus modulating the glottal airflow and generating sound waves
3. those waves then propagate into the vocal tract. The latter behaves as an acoustical resonator whose resonances are controlled by the positioning of the articulators (jaw, tongue, lips, velum).

These resonances determine the spectral content of the radiated pressure, creating peaks, called formants. These formants enable the auditory system of listeners to perceive and identify the different vowels of an articulated speech. Human oral communication is thus partly based on the control of resonances by the motion of the articulators. These resonances can also be used for artistic purposes in singing, especially in the high soprano range where pitch can no longer be controlled by the vocal folds alone.
In practice, however, characterizing the frequencies and bandwidths of vocal tract resonances remains difficult. Several methods exist:
•Formants analysis, the main tool currently used to study speech production, but reliable only at low fundamental frequencies.
•Magnetic Resonance Imaging which is an experimentally and computationally expensive method that moreover requires many assumptions to infer acoustic properties from the partially observed vocal tract geometry.
•Direct acoustic response methods, where a source generates waves at the lips and a simple microphone (with signal processing) records the response. This recent approach provides a good compromise between accuracy and intrusiveness

More information on the link below.