Audio editing and music practice tools

Audio Trimmer trims, cuts, fades, normalizes, resamples and changes the speed of MP3, WAV, M4A, FLAC and OGG audio on a waveform, and takes the audio track out of MP4, MOV and WebM video. It exports WAV (16-bit or 32-bit float) or MP3 and runs entirely in your browser, with nothing uploaded.

1. Open a recording or a supported video’s audio track.
2. Select the section on the waveform, then set cuts, fades, level or speed adjustments.
3. Export WAV or MP3 and listen to the beginning and ending to check the edit.

The tool decodes audio on your device. Keep an original copy, and avoid raising the level enough to clip the signal.

Metronome clicks from 20 to 400 BPM, timed on the audio clock, with per-beat accents, subdivisions, odd meters and their groupings, tap tempo and an optional counting voice. Its practice tools are a speed trainer, gap clicks, random drops, polyrhythms, a count-in and stop-after, and songs can be saved as a setlist in your browser.

1. Set a tempo from 20 to 400 BPM or use tap tempo.
2. Choose the meter, accents and subdivisions, then start at a comfortable pace.
3. Use the speed trainer or gap clicks for a focused exercise, and save songs to a setlist if useful.

Start below your performance tempo. Accents and subdivisions describe the pulse; increasing BPM is only one way to make practice harder.

Tuner listens through your microphone and shows the nearest note and how many cents sharp or flat it is, on a needle or a strobe, chromatically or for guitar, bass, ukulele, violin, viola, cello, double bass, mandolin and banjo. It reads 25 Hz to 4.2 kHz, lets A4 be set from 400 to 480 Hz and plays a reference drone; the microphone is never recorded or sent.

1. Allow microphone access and choose chromatic mode or an instrument tuning.
2. Play one string or note clearly and let the pitch reading settle.
3. Adjust toward the target note and center the cents reading; use a reference drone if helpful.

Background noise and multiple ringing strings can confuse pitch detection. Confirm the intended octave as well as the note name.

Voice Studio clones a voice you have permission to use from a 3 to 30 second recording or upload, then reads up to 800 characters of your text aloud in it with the Qwen3-TTS model, in 10 languages, as a WAV file. Unlike most qpkit tools it does not run in the browser: the recording and the text go to a Hugging Face GPU Space that makes the speech, and qpkit does not store the audio.

1. Record or upload a 3–30 second voice sample you have permission to use.
2. Enter up to 800 characters of text and select the appropriate language.
3. Generate speech, listen to the result and download the WAV file.

This tool sends the voice sample and text to a hosted Hugging Face GPU Space. It is an exception to qpkit’s mostly local processing; hosted availability can vary.