How a 30-Line Heuristic Fixes Beat-Tracking Algorithms Getting BPM Wrong
Popular beat-tracking tools like Essentia and Librosa frequently misidentify the tempo of modern pop and trap-influenced tracks, reporting half or double the actual BPM due to ambiguity in the metric hierarchy. The Weeknd's Blinding Lights, for example, is widely accepted at 171 BPM but is routinely returned as 85 BPM by these algorithms. The error occurs because beat trackers analyse onset envelopes and tempograms, and for tracks with halftime drum patterns, both the true tempo and its half-value appear as valid mathematical peaks. Spotify's now-deprecated audio_features endpoint suffered the same flaw, causing entire genres to be catalogued at incorrect tempos. A developer has published a 30-line heuristic that cross-checks reported BPM against onset rate, energy, and danceability signals to detect and correct these half-time and double-time errors.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in