Overtone singing, particularly the Tuvan sygyt style, achieves selective harmonic amplification with formant bandwidths on the order of 20-40 Hz and neighbor-harmonic attenuation exceeding 35 dB. The foundational work of researchers including Levin and Edgerton (1999), Adachi and Yamada (1999), Kob (2004), Saus (2009), and Bergevin et al. (2020) has established formant convergence as the primary acoustic mechanism. However, a bandwidth of 20-40 Hz at 1500 Hz implies an effective Q-factor of 40-75, values that may exceed what passive soft-tissue resonators typically achieve (Q on the order of 10-20, or up to 15-40 under optimized conditions). The hypothesis is most strongly motivated by the extreme end of this range. This discrepancy suggests that purely passive linear models, while descriptively accurate for frequency selection, may not fully account for the energetic profile (bandwidth and amplitude).This paper proposes a hybrid linear-nonlinear framework integrating three components: (1) a supra-lingual Helmholtz cavity with inter-molar lateral orifices as a previously uncharacterized anatomical structure; (2) vortex-acoustic coupling (lock-in) providing regenerative, nonlinear energy transfer that offsets tissue damping; and (3) a multi-parameter control framework explaining performer variability and skill progression. Demetrio Stratos (1978) demonstrated "flautofonia", flute-like sounds via supraglottic resonance without vocal fold vibration. This hypothesis proposes a complementary mechanism: analogous flow-acoustic coupling preserving modal phonation, achieving spectral focusing through vortex lock-in rather than replacing the glottal source. This framework generates testable predictions amenable to empirical investigation.

v2 (March 2026): Added documentation of pedagogical "sing to the side" practice from the overtone singing literature (Cosi & Tisato, 2003), providing published evidence for the lateral asymmetry prediction. Integrated transient (timbron) theory framework (Chen, 2016; Chen & Miller, 2020) to sharpen the Q-factor paradox and generate a new time-domain prediction (Prediction 8.5). Expanded preliminary behavioral validation with pilot Praat measurements. Added Section 9.5 reporting post-publication correspondence with researchers regarding imaging feasibility and practitioner confirmation of unilateral technique. Corrected pedagogical attribution in Section 6.1 (removed misclassification of non-overtone-singing researchers). Added AI assistance disclosure statement.