Core Vocal Metrics
The raw acoustic features that drive pitch, clarity, and consistency.
F0 (Fundamental Frequency)
The raw measurement of vocal pitch in Hertz (Hz) — the rate of vocal fold vibration. Forms the basis of the overall Pitch score.
f0_hz
: Average fundamental frequency of speech.f0_stability
: Consistency of pitch over time (feeds Stability score).
Formants (F1–F4)
Resonant frequencies of the vocal tract that shape vowel quality, clarity, and projection.
f1_hz
: First formant frequency.f2_hz
: Second formant; used in Tone Brightness.f3_hz
: Third formant frequency.f4_hz
: Fourth formant frequency.fd_hz
: Average distance between formants (articulation quality).
Harmonic-to-Noise Ratio (HNR)
Ratio of periodic (clean) sound to noise (breathiness/hoarseness). Higher is clearer.
hnr_db
: HNR in decibels; contributes to Clarity.
Shimmer
Short‑term amplitude variation. Lower shimmer indicates smoother, more consistent volume.
shimmer_db
: Amplitude fluctuation in dB (Volume Consistency input).
Jitter
Cycle‑to‑cycle pitch variation. Lower jitter sounds steadier and more controlled.
jitter_percent
: Pitch fluctuation percentage.
Voice Breaks
Moments where voicing stops or becomes unstable, often reflecting strain or hesitation.
voice_breaks_percent
: Portion of recording containing breaks.
Confidence and Reliability
Internal quality estimates that indicate how trustworthy the measurements are.
Overall Confidence
Composite control/stability score combining multiple indicators.
overall_confidence
: 0–1 confidence of overall vocal control.
Stability Confidence
How consistent pitch remained across time.
stability_confidence
: 0–1 steadiness of pitch.
Clarity Confidence
Reliability of clarity‑related measurements.
clarity_confidence
: 0–1 clarity quality estimate.
Consistency Confidence
Overall delivery consistency across the recording.
consistency_confidence
: 0–1 temporal consistency.
Measurement Reliability
Estimate of analysis reliability given audio conditions.
measurement_reliability
: 0–1 robustness of analysis.
Sample Quality
Overall capture quality; higher is cleaner audio with fewer distortions.
sample_quality
: 0–1 audio quality indicator.