Extended Vocal Techniques Spectrum (Music)

Overtone Singing

Structured resonances, controlled transitions, glissandi, and dynamic variations for expressive vocal synthesis.

PREVIEW

Extended Vocal Techniques Spectrum (Music)

This dataset is currently in production. Preview audio and full specifications will be added as they become available.

This is a preview release provided via Hugging Face for evaluation purposes. Initial recording passes are available to assess capture quality, labeling structure, and dataset relevance. The full dataset will be delivered privately with complete Proteus provenance, integrity, and documentation upon licensing.

This dataset is complete and available for licensing. A public preview subset will be released via Hugging Face on a rolling basis.

Overtone singing is a vocal technique in which a performer sustains a single fundamental pitch while selectively amplifying upper harmonics through precise shaping of the vocal tract. Rather than producing multiple pitches mechanically, the technique exploits the natural harmonic structure of the human voice, allowing individual overtones to emerge as distinct perceptual elements above the fundamental tone.

‍

Acoustically, overtone singing provides a rare and highly controlled window into vocal tract resonance behavior, spectral shaping, and harmonic emphasis. Expressivity arises through fine-grained adjustments of tongue position, mouth shape, and breath control rather than changes in pitch or amplitude. These characteristics make overtone singing particularly well suited for articulation-level analysis and modeling, where disentangling source excitation from resonant filtering and gesture-driven spectral movement is essential.

‍

Audio Demonstrations

A three-part listening benchmark: a mixed musical demo built from this dataset, the raw source clip, and an AI model’s attempt to reproduce the same prompt.

PRODUCED REFERENCE

A musical demonstration created by replacing a state-of-the-art AI-generated lead instrument with original source recordings from this dataset, then arranged and mastered to preserve musical context. This approach allows direct comparison between current-generation model output and real, rights-cleared acoustic source material.

RAW DATASET CLIP

Directly from the dataset: an isolated, unprocessed example of the source recording.

AI MODEL BENCHMARK
AI MELODIC PROXY (Overtone Singing Unsupported)

An unmodified output from a current-gen AI model given the same musical prompt. Included to illustrate where today’s systems still differ from real, recorded sources.

An unmodified output from a current-generation AI music model given the same melodic prompt. Because current generative music models do not support overtone-based spectral melody, this example substitutes a conventional high-pitched melodic instrument as a proxy. The result illustrates a structural limitation of today’s systems when attempting to approximate overtone singing techniques.

AI model approximations generated using publicly available state-of-the-art music generation systems.