drift3

pitch trace exploration tool

• View source on Github
• Install a DMG for Mac (requires Gentle and macOS Mojave 10.14.3)
• Try the demo online

features

• compute prosodic measures on transcribed audio files in their entirety or in parts. copy data directly to excel.
• interactive playback with f0 pitch visualization. compare different files side by side.
• download data as JSON (alignment), CSV (pitch/alignment/phonemes), TXT (pitch), or MATLAB (prosodic measure, and the rest)

notes and caveats

Please contact Robert M Ochshorn or file issues on Github with bug reports and feature requests.

While the Ellis pitch-tracking algorithm that Drift uses is state-of-the-art, and works well even on noisy recordings, some errors in pitch-tracking are inevitable.

See “Twelve Prosodic Measures” in “Beyond Poet Voice: Sampling the Non-Performance Styles of 100 American Poets” for ideas of how to analyze the pitch and timing data Drift provides.