One upload. Every audio tool you need.

Separate any song into two clean tracks: the vocal on its own and the full instrumental behind it.

Break a song into four separate tracks (vocals, drums, bass and everything else), ready to remix or remaster.

Separate a song into six tracks (vocals, drums, bass, guitar, piano and other) for detailed remixing.

Removes the lead vocal so you can sing along, and hands you the isolated vocal too.

Separates a film or video soundtrack into three tracks: dialogue, music and sound effects.

Removes background hiss, hum and room noise from spoken-word recordings, leaving the voice clear.

Reduces broadband hiss and noise across a full music mix without hollowing out the instruments.

Takes the room out: reduces reverb and echo so a recording sounds close and dry instead of distant.

One-pass repair for voice: removes noise, sharpens clarity and rebuilds missing high frequencies.

Fixes recordings that distorted from being recorded too loud, rebuilding the damaged detail.

Rebuilds detail crushed out by low-bitrate MP3s and other lossy files, for a fuller, less brittle sound.

Tell it what's wrong in plain words ('too muddy', 'remove the hiss') and it restores and masters to match.

Reduce clipping, short dropouts and compression damage in calls and remote speech recordings.

AI upscaling that reconstructs crisp high-frequency detail for dull audio. The best quality, but slower.

Rebuilds the high frequencies missing from dull, muffled or heavily-compressed recordings for a fast fix for lifeless audio.

A near-free brightener tuned for the human voice that makes muffled or telephone-quality speech clear and present.

Writes out a song's lyrics with a timestamp on every word, ready for captions, karaoke or subtitles.

Listens to a track and suggests genre, mood and instrument tags, handy for sorting and labelling a library.

Reports exactly how loud your audio is against streaming and broadcast standards, without changing it.

Detects the tempo and marks every beat and downbeat, giving you the grid you need to sync, loop or remix in time.

Finds the musical key of a track and maps out its chord progression over time.

Finds where the sections of a track begin and end (intro, verse, chorus, drop), and groups the similar ones.

Turns a melody or instrument recording into editable MIDI notes you can open in any music software.

Turn a podcast or interview into a timed transcript with speaker labels and TXT, SRT and JSON downloads.

Sketch Mandarin lyrics and melody as an editable score. Timing is approximate.

Makes one recording sound like it was performed by a different voice you provide, whether speaking or singing.

Nudges an off-pitch vocal toward the right notes with subtle correction by default, or hard-snap if you want the effect.

Turns written text into natural-sounding spoken narration, with a choice of voices. No recording needed.

Give it a short clip of any voice and it speaks your text in that voice. Every result is watermarked.

Transfer a reference singer’s voice onto your isolated vocal take while keeping its words and melody.

Describe a style and it writes and produces original music. Add your own lyrics or leave it instrumental.

Continues a track past its end (or before its start), or regenerates a chosen section in place.

Create backing music that follows the rhythm and harmony of your isolated vocal.

Try a new style for your music or generate an instrument track around your existing clip.

Replace a short passage or add a new ending, guided by your description.

Change the sound of a selected passage while using the original as a starting point.

Sets your track to the loudness streaming platforms expect, so it doesn't play back too quiet or too loud.

Point it at a song whose sound you love and it makes your track match that tonal balance and loudness.

Pushes overall level up to a safe ceiling without letting peaks distort as the final loudness stage of a master.

Makes a track longer or shorter without changing its pitch, so it stays in the same key, just faster or slower.

Moves audio up or down in pitch by any number of semitones without changing its speed or length.

Convert audio between MP3, AAC, Opus, FLAC and WAV, with control over sample rate and quality.

Cut audio down to just the part you want, with optional fade-in and fade-out at the edges.

Add a smooth fade-in at the start or fade-out at the end, with a choice of fade shapes.

Speed audio up or slow it down like a record: the pitch rises and falls with the speed. From 0.1× to 8×.

Turn a track up or down by a set amount, or normalize its loudest peak to a target level.

Shape the tone by boosting or cutting bass, treble and any frequency between, with presets or precise bands.

Even out volume, silence the noise between phrases, or tame harsh "s" sounds with simple presets.

Automatically trims silent gaps: the dead air at the start, the end, or throughout a recording.

Convert between mono and stereo, swap or isolate the left/right channels, or do mid-side encoding.

Play a clip backwards for reverse-cymbal swells, hidden-message effects or creative transitions.

Pull the soundtrack out of a video file and save it as audio in formats including MP3, FLAC, WAV and more.

Read or set a file's title, artist, album and other tags, and embed the cover-art image.

Render a picture of a track's frequencies over time, useful for spotting noise, cutoffs or fake quality.

Render the classic waveform picture of a track (the shape of its volume over time) as a PNG.

The DJ 'power-down' effect: audio slows and drops in pitch to a halt, like a tape or turntable stopping.

Builds tension for a drop with a repeat that speeds up and rises in pitch into a climax.

The come-down counterpart to the riser: a repeat that slows and drops in pitch to release energy.

Chops a sustained sound into a rhythmic on-off pulse in time with the beat, creating the classic trance-gate effect.