In-browser acoustic metrics for speech — arousal and expressiveness, computed locally from pitch, loudness, and rhythm.
One line, re-rendered at rising expressiveness — 31% → 73%.
One line, re-rendered at rising emotional intensity — 22% → 53%.
…or drag & drop wav / mp3 here. Everything stays in your browser.
arousal blends loudness, pitch variation, and speaking rate — higher reads as more energetic delivery.
The technical range is 0–1, but in practice, anything below 20% or above 60% is probably an outlier.
expressiveness blends pitch range, loudness variation, and pitch variation — higher reads as more dynamic, less flat.
The technical range is 0–1, but in practice, anything below 20% or above 80% is probably an outlier.
Pitch is tracked with probabilistic YIN, rhythm from onset-strength peaks, loudness from framewise RMS. Click any column header to sort.