# Vocal Color > Record your voice and discover your unique color. Vocal Color is a free web app that analyzes your voice and translates a five-dimensional acoustic profile into a color that is uniquely yours. ## What it does You record yourself reading a short standardized passage based on the Stella passage ("Please call Stella. Ask her to bring these things with her from the store..."). Vocal Color analyzes the recording entirely in your browser. Your audio is never uploaded or stored. The app derives a voice result from five primary dimensions: - Pitch: Estimated fundamental frequency, often called F0, measured in Hz across voiced speech. It captures how high or low the voice sits within a calibrated speaking range. - Timbre: Resonance and spectral shape created by the vocal tract. It is derived from multiple resonance and spectral features that contribute to the voice's tonal character. - Clarity: The balance of periodic harmonic energy versus breathy or noisy energy. It is derived from phonation and harmonicity features. - Rhythm: Speech timing and energy patterning over time, including pacing and pulse structure. It is derived from temporal features. - Flow: How continuous or segmented phrasing sounds, from smoother legato delivery to more separated staccato delivery. It is derived from continuity and transition features. These measurements are used to derive a vocal color. Pitch sets the base color path, timbre, clarity, rhythm, and flow apply additional adjustments, and the final result is rendered into display-safe outputs such as hex and HSL. The analysis is octave-aware, so similar relative positions within an octave can echo similar hue regions while the full voice profile still shapes the final color. ## Repeatability Results are designed to be repeatable under stable recording conditions. In a quiet room, with the same device, similar mic distance, and a similar speaking style, repeated takes should stay in the same general color neighborhood and profile shape. Small shifts between takes are normal. Large shifts usually reflect noise, mic placement, vocal fatigue, or unusually expressive delivery. ## What Is Measured vs What Is Interpreted Vocal Color measures acoustic properties of the recorded voice signal. It does not directly measure personality, identity, intelligence, attractiveness, or health. The measured layer includes pitch, timbre, clarity, rhythm, and flow. The interpreted layer is the mapping from those measurements into color, labels, and descriptive language. The signal analysis is empirical. The meaning attached to the color is a creative layer built on top of the measured voice features. ## Output Your result includes: - A named color result - A unique hex color code - HSL values - A voice profile across five dimensions: Pitch, Timbre, Clarity, Rhythm, and Flow - Listener-oriented perception insights for each dimension including possible strengths and things to watch out for - Numerical assessment of each dimension - A shareable image card - Recording-quality guidance when the sample is noisy or unstable ## How Color Should Be Understood The color is best understood as a compact aesthetic summary of measurable vocal behavior. It is relative across a calibrated space of human voices. Similar colors suggest similar measured patterns, not identical voices or identities. Displayed hex and HSL values are presentation outputs of the mapping, not direct physical measurements like Hz or dB. ## Sensitivity - The system is most sensitive to stable pitch extraction, consistent voicing, room noise, microphone distance, and vocal condition. - The final color is derived from a weighted combination of all five dimensions, with pitch assigned the strongest influence and timbre, clarity, rhythm, and flow contributing secondary adjustments. - The system will respond to changes in delivery, effort, and recording quality. ## Repeated Use Guidance - For self-comparison, use the same device, room, mic distance, and reading style whenever possible. - If tracking change over time, record two or three clean takes per session and compare the cluster rather than a single take. - Repeated use is best for observing ranges and patterns, not for chasing one exact hex code. ## Sensitive Use Guidance - Vocal Color should not be used for diagnosis, hiring, admissions, truth testing, identity verification, or judgments about a person's intelligence, identity, gender, race, health, or worth. - If the recording is noisy or unstable, the result should be treated as low-confidence and not overinterpreted. ## Privacy - 100% client-side analysis - No backend voice processing - No audio uploads - No audio storage ## Tech - Built with React and Vite, deployed on Vercel - Uses browser microphone access and local audio decoding - Custom DSP pipeline including pitch detection, FFT and resonance analysis, phonation clarity measures, temporal speech-pattern features, and derived voice-feature mapping - Uses Vercel Analytics for aggregate product analytics; no voice recordings are collected ## Audience - People interested in self-discovery, identity tools, and creative personal analysis - Voice professionals curious about an aesthetic interpretation of vocal qualities - Anyone who wants to know what color their voice might be ## What it is not - Not a medical or clinical tool - Not a biometric identity or verification system - Not a scientifically validated personality test - Not a permanent judgment of a person or voice - Not AI-powered. The entire analysis pipeline uses deterministic signal processing, not machine learning or trained models The color system is a creative interpretation of real acoustic features. Results can shift between takes based on room noise, microphone distance, and vocal condition. ## URL https://www.vocalcolor.net ## Related pages - https://www.vocalcolor.net/how-it-works.html (how the voice analysis and color mapping work) - https://www.vocalcolor.net/faq.html (common questions about measurement, variation, and device support) - https://www.vocalcolor.net/privacy.html (privacy, analytics, and data handling)