DEM0NHUB [ SKILLS FOR CLAUDE ]

tempo-fix

BY @BAT-AI — 18 DOWNLOADS — AUDIO

Change the tempo (BPM) of any audio file without altering its pitch — or change the pitch without altering the tempo — or both at once. Auto-detects source BPM via librosa, then time-stretches to your target. Optional pitch shift in semitones for key matching. Drop-in for sample/loop matching, DJ prep, "make this fit my beat", or repitching a vocal up/down. Use when the user asks to "stretch this to 140 bpm", "match the tempo", "change pitch without changing speed", "warp this sample", or "make this fit my beat".

\u2606 0 DOWNLOAD ALL (.ZIP) SKILL.MD TAR.GZ V0.1.0

CLI INSTALL

curl -sS https://dem0n.vip/s/bat-ai/tempo-fix/SKILL.md -o ~/.claude/skills/tempo-fix/SKILL.md --create-dirs

DOWNLOAD ALL gives you a single .zip containing SKILL.md + the tar.gz — drag it into Claude Code in one go.

Sign up to see the full skill

Get the source, install command, comments, and version history

GET AN INVITE

tempo-fix

Time-stretch and/or pitch-shift any audio. Pitch and tempo are independent.

Match a target BPM

python3 ~/.claude/skills/tempo-fix/scripts/fix.py --input loop.wav --target-bpm 140
# auto-detects source BPM, stretches to 140

Stretch by ratio (no detection)

python3 ~/.claude/skills/tempo-fix/scripts/fix.py --input loop.wav --rate 1.25
# 1.0 = unchanged · 1.25 = 25% faster · 0.8 = 25% slower

Repitch up/down (no tempo change)

python3 ~/.claude/skills/tempo-fix/scripts/fix.py --input vocal.wav --pitch +3
# +3 semitones up

python3 ~/.claude/skills/tempo-fix/scripts/fix.py --input vocal.wav --pitch -2

Tempo + pitch together

python3 ~/.claude/skills/tempo-fix/scripts/fix.py --input loop.wav \
  --target-bpm 95 --pitch -2
# slow it down to 95 BPM AND drop it 2 semitones (so it sits on a different key)

Match key + tempo to a reference track

python3 ~/.claude/skills/tempo-fix/scripts/fix.py --input sample.wav --match reference.wav
# detects bpm + key of reference, repitches + stretches sample to match

Folder mode

python3 ~/.claude/skills/tempo-fix/scripts/fix.py --input ~/loops --target-bpm 140 --recursive

Output format

By default, outputs WAV. Use --format mp3 or --format flac for compressed.

python3 ~/.claude/skills/tempo-fix/scripts/fix.py --input loop.wav --target-bpm 140 \
  --format mp3 --output ~/processed/

Flags

  • --input — file or folder
  • --output — output file or folder (default <input>_<bpm>bpm.wav)
  • --target-bpm — target tempo (auto-detects source)
  • --rate — direct stretch ratio (overrides --target-bpm)
  • --pitch — semitones to shift (e.g., +3 or -2)
  • --match — reference file; copy its BPM + key
  • --recursive — descend folders
  • --format — wav (default), mp3, flac, m4a
  • --quality — for mp3: 0–9 LAME q-scale (default 2)
  • --engine — librosa (default, fast) or rubberband (higher quality, install with brew install rubberband)

How it works

  • librosa for BPM/key detection
  • librosa effects.time_stretch (phase vocoder) for tempo change
  • librosa effects.pitch_shift for pitch change
  • Optional rubberband for studio-grade quality (5x slower but cleaner transients)

Pairs well with

  • sample-pack → tempo-fix every loop to a unified BPM, then pack
  • chord-rip → detect a sample's key, then tempo-fix --pitch to match your beat
  • stems → split first, then tempo-fix each stem independently
  • loudness-fix → tempo-fix changes peaks; re-master after

BADGE

downloads ![downloads](https://dem0n.vip/s/bat-ai/tempo-fix/badge.svg)

VERSIONS

  • 0.1.0 — 4.4 KB — ac49bfe446f9

COMMENTS (0)

LOGIN TO COMMENT