Human Vocality Primitives
Whisper phonation with varying pressure, turbulence, and vowel shaping.
Whisper and aspiration are airflow-driven vocal behaviors produced without sustained vocal fold vibration, relying instead on controlled breath, glottal configuration, and vocal tract shaping to generate sound. In these techniques, acoustic energy is created primarily through turbulent airflow rather than harmonic excitation, resulting in noise-shaped vocal output with no stable fundamental pitch. Expressive control emerges through subtle modulation of airflow intensity, mouth shape, articulatory configuration, and transitions into and out of silence rather than changes in loudness or pitch.
Acoustically, whisper phonation and aspiration occupy a broadband, noise-dominant timbral space characterized by diffuse spectral energy, formant-shaped filtering, and fine-grained temporal variation. Expressivity is conveyed through airflow texture, articulatory gesture timing, and the interaction between breath noise and the resonant properties of the vocal tract. These characteristics make whisper and aspiration particularly well suited for modeling unvoiced vocal behavior, breath-aware speech systems, and airflow-centric sound analysis, where precise control of onset behavior, noise structure, and repeatable articulatory gestures is essential.
Key technical details for this dataset — including file counts, duration, delivery format, and session context.
Planned technical specifications and recording standards for this dataset.
Total Files:
126
Total Files (Preview):
Sample Rate (Hz):
96000
Bit Depth (Delivery):
24
Dataset Version:
v1.0
Recording Environment:
Treated Studio
Microphone Configuration:
Rode NT1-A positioned 4-5 inches from mouth
Performer:
Blake Pullen
Recording Dates:
Jan. 2nd 2026
Recording Location:
Las Vegas, NV
Produced using standardized capture, editing, and QC protocols with versioned metadata and Proteus-backed provenance.
This dataset includes a structured set of whisper and aspiration-based vocal articulations, recorded in isolation to support articulation-aware modeling and analysis of unvoiced vocal behavior.
Articulations include:
• Sustained whisper phonation at varying airflow intensities
• Fricative-based whisper gestures produced through controlled articulatory shaping
• Mouth-shape and constriction variations affecting airflow texture
• Controlled onsets, offsets, and transitions into and out of silence
• Aspiration-dominant airflow events without sustained phonation
Articulations are recorded without linguistic content, accompaniment, or rhythmic framing to preserve clarity, separability, and modeling utility across speech, voice synthesis, and airflow-centric analysis contexts.
Captured with expert musicians and vocalists across global traditions — ensuring each dataset carries authentic nuance, human expression, and rights-managed provenance.

Blake Pullen is a trained vocalist, instrumentalist, and recording engineer, and the founder of Harmonic Frontier Audio. He performs and records HFA’s initial datasets to set the capture and labeling standard for the catalog.
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.
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.
Directly from the dataset: an isolated, unprocessed example of the source recording.
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.
AI model approximations generated using publicly available state-of-the-art music generation systems.
Harmonic Frontier Audio datasets are licensed directly to research teams, startups, and enterprise partners. Access models and terms vary based on use case, scale, and integration needs.
All datasets are delivered with versioned metadata, documented workflows, and Proteus-backed integrity manifests.
We typically respond to inquiries within 1 day.
This dataset is actively being recorded and prepared. You can request early access, previews, or discuss licensing timelines.
We typically respond within 1–2 business days.