Free audio tools: split songs, clean voices, cut and convert

• Vocal RemoverStemGrabSplits a song into an instrumental and a vocal track; on most modern studio mixes the result is clean, and live, dense or heavily reverbed recordings leave some residue.
• Acapella ExtractorStemGrabExtract an acapella from a song you upload, free, with no watermark and no account; harmonies stay in the vocal, and live or very reverbed recordings come out rougher.
• Extract Vocals from SongStemGrabPull the isolated vocal out of a song you upload as a 192 kbps MP3, plus the instrumental; it works best on well-mixed studio recordings and keeps backing vocals together with the lead.
• Remove Background Noise from VideoCleanGrabReduce steady background noise in a video's audio; you get the same video back with the picture copied untouched (other audio and subtitle tracks are not kept).
• Split Song into TracksStemGrabOne upload gives vocals, drums, bass, guitar, piano, other and an instrumental, with no account needed.
• Remove Background Noise from AudioCleanGrabRemove steady background noise (hum, hiss, fans, traffic, keyboard) from a recording of speech; the voice stays natural at Balanced, and Strong can thin it.
• Audio CutterStemGrabCut an MP3 online: drag the two handles over the waveform, press cut, and download the result.
• Extract Audio from VideoStemGrabExtract the audio from an MP4, MOV, MKV or WEBM video as an MP3, free, up to 120 MB or an hour; a few unusual containers cannot be read.
• Remove Drums from a SongStemGrabSplit a song and get the drums as their own MP3, free; to remove the drums, mix the other stems back together in an audio editor.
• Instrumental MakerStemGrabStrip the vocals from a song and get the instrumental plus the isolated vocal as 192 kbps MP3 files; WAV and the individual drums, bass, guitar and piano stems are in the main splitter.
• Karaoke MakerStemGrabMake a karaoke instrumental from a song you upload; the lyric video is built from automatically written lyrics, so check and fix the lines before you render the MP4.
• MP3 to MIDIStemGrabVocals, bass, guitar and piano each come back as their own track, plus one combined file with every part separated; expect to fix some notes by hand.
• Remove Background Music from VideoStemGrabSeparate speech from the music in a video or audio file; you get the voice and the music as separate audio files and put the voice back under your picture in an editor.
• Pitch ChangerStemGrabShift a track up or down by semitones, or slow it down or speed it up without moving the pitch; upload an MP3, WAV or video file (up to 10 minutes) and get an MP3 back.
• Remove Hum and Hiss from AudioCleanGrabTurn down steady electrical hum and background hiss in a recording of speech; Balanced keeps the voice natural, and very heavy hum removal can slightly affect low tones.
• Remove Room Echo from AudioCleanGrabReduces room echo and reverb in a voice recording rather than removing it completely.
• Remix StemsStemGrabSplit a song you upload into vocals, drums, bass, guitar, piano and other as separate 192 kbps MP3 files (WAV is in the main splitter); guitar and piano can bleed on dense mixes.
• Voice CleanerCleanGrabFree: tidies up a voice memo, interview or call, so the speaker comes back clearer with the room noise pulled down.
• Silence RemoverStemGrabThis tool removes long silences and dead air from a podcast or voice recording automatically, free: no account, no watermark, nothing installed.
• Isolate Bass from SongStemGrabIsolate the bass of a song as its own MP3 track, free. To hear the song without it, leave that file out when you mix the other stems in an editor.
• Chord FinderStemGrabUpload a song and get its chords bar by bar, lighting up as it plays; treat it as a first draft and check by ear.
• Text to SpeechScribeGrabType your text and get a voice back: 5,000 characters per go, 10 minutes of speech a day, no account.
• Audio JoinerStemGrabJoins two or more audio files, even in mixed formats, into a single MP3 that you can download.
• Song Key ChangerStemGrabReads the key off your file, shifts it up or down in an MP3, WAV or FLAC, and tells you the key you land in.
• BPM FinderStemGrabRead a song's tempo in beats per minute in about two seconds, with its key and Camelot code; it can land on half or double the real tempo, so check it against your own count.
• Enhance Speech OnlineCleanGrabDrop in a recording and get a clearer voice back from an AI speech enhancer. No account needed.
• Remove Noise from RecordingsCleanGrabTurn steady background noise (hum, hiss, fans, traffic) down in a recording of speech; the voice stays natural at Gentle and Balanced, and Strong can thin it.
• Remove Wind Noise from AudioCleanGrabReduces steady noise in a video or voice recording shot outside; the tool's own page reports that the strongest setting lowered synthetic wind noise by only about 0.4 dB, so expect little change. A foam windscreen on the microphone helps far more than software.
• Remove Fabric Rustling Noise from AudioCleanGrabBuilt for steady noise, so short rustles from a lavalier or headset mic may only be turned down; I have not measured how much.
• Voice RecorderStemGrabRecords your voice in your browser with nothing to install and no account to create.
• Key FinderStemGrabFind the key, scale, Camelot code and BPM of a song in about two seconds; it is most reliable on well-produced music and can confuse a key with its relative major or minor.
• Clean Audio for YouTube VideoCleanGrabClean the noise out of the audio of a video before you upload it to YouTube; you get the same video back with the picture untouched, plus the cleaned MP3 and WAV.
• Remove Sibilance from Voice RecordingsCleanGrabI do not de-ess: this removes steady background hiss and mic static from a voice recording, and sharp S and T sounds in the voice stay exactly as recorded.
• Audio EnhancerCleanGrabMake a voice recording clearer, with background noise, hum and hiss pulled down.
• Practice Any Instrument With StemsStemGrabA guide to practicing along with a song: split it into vocals, drums, bass, guitar, piano and other, then mute your own part in the results mixer or your editor.
• Voice EnhancerCleanGrabA memo, call, interview or lecture comes back clearer, with the background noise pulled down.
• Clean Up a Zoom or Teams RecordingCleanGrabYou get a cleaned MP3 and WAV, plus the same video with the cleaned sound if you upload a video; steady noise is turned down, overlapping voices are not separated.
• Podcast Audio CleanerCleanGrabTurn down hum, hiss, fans and traffic in a podcast episode before you publish it; it works on one mixed file, not on separate speakers.
• Song Key and BPM FinderStemGrabFind a song's key and BPM in one pass, with the Camelot code; check borderline results by ear.
• Isolate Voice from Background NoiseCleanGrabTurn down barking and birdsong behind a speaker with Voice only mode, in audio or video; it cannot pick one voice out of several people talking.
• Change Audio VolumeConvertNinjaTurns an MP3, WAV or M4A file louder or quieter by a set number of decibels.
• Compress AudioConvertNinjaShrink an audio file into a smaller MP3 (about 190 kbps variable bitrate), free, up to 20 minutes; a file that is already small is shrunk a little further or, if that cannot make it smaller, refused.
• Fade AudioConvertNinjaFade a track in or out. Works on an MP3, a music track or a sound effect.
• Normalize AudioConvertNinjaBrings an audio file to a standard loudness so that quiet and loud parts even out.
• Remove Dog Barking from VideoCleanGrabTurns barking down in a video or audio file with the Voice only mode; in a small test it came down by 14 dB to completely gone, about 44 dB on average, so listen before you publish.
• Remove Silence From Audio (Automatic)ConvertNinjaCut the silent gaps out of a recording automatically; for speech, StemGrab's Silence Remover has a sensitivity slider.
• Reverse AudioConvertNinjaPlays an audio file backwards and lets you download the reversed result as an MP3.
• Tap TempoStemGrabTap a button or the spacebar and read the tempo live, averaged over your last taps.

A few jobs are smaller and quicker: cutting or joining clips, changing the volume, fading in and out, reversing a sound, finding the key and tempo of a song, tapping a beat. If all you need is a trim, start with those instead of a separation, which takes longer and uses a graphics card.

Separation needs a mix to work with. A finished studio mix gives it the most to work with; a phone recording of a live room has the room mixed into everything, so expect some of it to come along. Noise removal works best on steady noise like hiss or a fan, and worst when somebody talks over another voice. Splitting a song allows 10 free jobs a day and noise removal 10, while cutting, joining, pitch, key and tempo have no daily counter. Your upload is deleted right after it has been processed.

A vocal remover splits a mix into vocals and the rest. The same split powers the Instrumental Maker and the Karaoke Maker, which keep the music and drop the singer. To practice a part, isolate voice or one instrument and turn the others down. A free split takes tracks of up to 15 minutes. If you are coming from LALAL AI or Moises, StemGrab has a guide for each: Free LALAL AI Alternative and Free Moises Alternative.

Remove background noise from audio before you edit or publish a voice recording, because later steps such as compression work better on a clean signal; you usually do not need it before you transcribe. The Audio Enhancer and the Podcast Audio Cleaner turn down hum and hiss and reduce room echo, but neither can rescue a voice buried under a second voice. Remove Sibilance from Voice Recordings is a name people search for, but it does not de-ess: it removes steady background hiss and mic static, and sharp S and T sounds stay as recorded. Remove Fabric Rustling Noise from Audio is built for steady noise, so short rustles may only be turned down. Use Normalize Audio only after the cleanup, so the level is raised for the speech and not for the noise. A free cleanup takes recordings of up to 45 minutes.

A voice recorder on a phone often produces quiet files that sometimes clip, so look at the waveform before you start. Recordings made in a small, soft room with the phone close to the mouth need much less repair than a recorder left running in a kitchen. You can also remove background music when a podcast bed is too loud under the speech, though a crowded mix keeps some of it.

The Audio Cutter is the quickest tool here: trim the silence from a voice memo or cut the start of a song for a ringtone, then use the Audio Joiner to put separate takes together. Extract Audio from Video gives you the sound when you only need that, and Compress Audio shrinks a file that is too large to send, and saves a smaller MP3 (about 190 kbps) with no bitrate setting to choose. Reverse Audio plays a file backwards for effects, or just to hear how a recording sounds that way.

MP3 to MIDI turns the melody of a recording into notes you can edit: it converts each separated part on its own, clean recordings give better notes, and you should expect to fix some by hand; the free limit is 7 minutes. Text to speech is the reverse of transcription: you type the words and get a voice back, which is useful for narration drafts and for hearing how a script sounds.

Song key and BPM are what a DJ or a musician checks before mixing. Key Finder reports the key, BPM Finder reports the beats per minute, and Song Key and BPM Finder gives both from one upload. If a song sits in the wrong key for your voice, Pitch Changer and Song Key Changer shift it up or down in semitones. Tap Tempo is for when you only have a feeling: tap along to the beat and read the BPM. Detection is most reliable on a track with a steady beat.

Other groups: Photo tools · Video tools · Transcribe & subtitles · PDF & documents · File converters · Text tools.