Free online audio tools
Two tools that independent musicians use every week: an AI mastering engine that gets a mix ready for release, and a stem splitter that takes a song apart into vocals and instruments. Both run online, both let you hear the result before you commit, and neither needs a subscription.
Audio converter
Convert between MP3, WAV and AIFF, or pull the audio out of an MP4, MOV, M4A, FLAC or OGG file. Runs in your browser, so the file is never uploaded. Dedicated pages for MP4 to MP3, MP3 to WAV, WAV to MP3 and M4A to MP3.
- No upload, no account, no queue
- MP3 at 64 to 320 kbps
- WAV at 16, 24 or 32-bit float, AIFF at 16 or 24 bit
- Sample rate from 8000 to 96000 Hz, mono or stereo
- Reads MP3, WAV, AIFF, FLAC, OGG, M4A, MP4 and MOV
For: Anyone who needs a file in another format: a voice memo as MP3, a mix as WAV for the session, the sound of a video as audio.
Tuner, bpm finder and metronome
Three practice tools: a chromatic tuner that shows notes as sargam if you want, a bpm finder that taps or measures a song, and a metronome that counts taals with sam, tali and khali.
- Tuner with Western or sargam names and a chosen Sa
- BPM by tapping or from a file
- Metronome with Teentaal, Keherwa, Dadra, Rupak, Jhaptaal, Ektaal
- Runs in the browser; nothing is recorded or uploaded
- Free, no account
For: Singers and instrumentalists practising, students learning taals, producers checking a tempo.
MP3 cutter and ringtone maker
Cut a piece out of any audio file on a waveform, add fades, and save it as MP3 or WAV. The ringtone maker page starts with a 30-second piece and the fades a ringtone wants.
- No upload, no account
- Start and end on a waveform, to a tenth of a second
- Fade in and out, play before saving
- MP3 at 64 to 320 kbps or WAV
- Reads MP3, WAV, AIFF, FLAC, OGG, M4A and MP4 audio
For: Anyone trimming a recording, making a ringtone or alarm sound, or pulling one section out of a long file.
Noise remover
Take the hiss, hum and room noise out of a voice memo, a riyaz recording or a demo. The tool learns the noise from the quiet parts of the file; compare with the original before saving as MP3 or WAV.
- No upload, no account
- Noise sample found automatically or from the first second
- Four strengths and a sensitivity control
- Hear the cleaned and the original before saving
- MP3 at 64 to 320 kbps or WAV
For: Anyone with a phone recording that has fan, traffic or microphone noise under the sound they want.
Audio joiner
Merge two or more audio files into one, in any order, with silence or a crossfade between them, and save as MP3 or WAV. Formats and sample rates can be mixed.
- No upload, no account
- Any number of files, in any order
- Silence or a crossfade between files
- Mixed formats and sample rates matched automatically
- MP3 at 64 to 320 kbps or WAV
For: Anyone stitching recordings together: parts of a take, chapters of a talk, a set of songs into one file.
Pitch changer
Shift a song or a backing track by up to 12 semitones, keeping the tempo or changing speed with the pitch. Preview 20 seconds before saving as MP3 or WAV.
- No upload, no account
- An octave up or down in semitone steps
- Keep the tempo, or tape-style speed change
- Preview before saving
- MP3 at 64 to 320 kbps or WAV
For: Singers moving a song into their range, students matching a track to their Sa, producers shifting a sample.
Audio compressor
Reduce the size of an MP3, WAV or M4A by re-encoding it at a chosen bitrate, sample rate and channel count, or fitted under a target size, with the expected size shown first.
- No upload, no account
- Fit under a target size or choose the bitrate
- Expected size shown before you compress
- 44100, 32000 or 22050 Hz, mono or stereo
- Reads MP3, WAV, AIFF, FLAC, OGG, M4A and MP4 audio
For: Anyone sending audio over chat or email, or storing a lot of recordings.
Karaoke maker
Turn any song into a karaoke track: the instrumental without the singing, plus the isolated vocal, as WAV files. Hear a 21-second preview before you process the full song.
- Instrumental karaoke track and isolated vocal as separate WAV files
- 21-second preview without an account
- MP3, WAV, FLAC and AIFF up to 100 MB
- Same AI model as the stem splitter (Demucs)
- Free, no subscription
For: Singers who want to practise or perform a song without the original voice, karaoke nights, cover recordings.
Vocal remover
Take the voice out of any song to get a karaoke track, or keep the voice alone as an acapella. Hear a 21-second preview before you process the full song.
- Instrumental and isolated vocal as separate WAV files
- 21-second preview without an account
- MP3, WAV, FLAC and AIFF up to 100 MB
- Same AI model as the stem splitter (Demucs)
- Free, no subscription
For: Singers practising and making covers, karaoke nights, remixers who need the acapella.
Online audio mastering
Upload a mix, choose a genre preset, set a loudness target in LUFS and preview the master before you download it as a WAV. Built for singles going to Spotify, Apple Music and YouTube.
- Genre presets and adjustable parameters
- LUFS loudness targeting with true-peak limiting
- Instant preview before download
- WAV output ready for streaming platforms
- Free, no subscription
For: Independent artists, producers and home studios who need a release-ready master without a studio booking.
AI stem splitter
Separate any song into vocals and instrumental, or into vocals, drums, bass and other. Hear a 21-second preview of the separation without signing up, then process the full track in the background.
- 21-second preview without an account
- 2-stem and 4-stem separation
- MP3, WAV, FLAC and AIFF up to 100 MB
- Stems as 44.1 kHz stereo WAV files
- Runs in the background; results in your dashboard
For: Singers making covers and practice tracks, producers remixing, DJs and teachers isolating parts.
How the tools work
Upload a file, hear a preview, then process the full track. Mastering returns a WAV master at the loudness you choose; stem separation returns each stem as its own WAV in your dashboard. Files are processed on our servers and you keep your rights to your own recordings.
Made by a label, for musicians
BeatRootRaga is an independent label that records ghazals, Indian classical and film-song covers. The tools grew out of the label's own work: mastering singles for streaming and separating reference tracks so singers can practise. That is why they favour natural dynamics over loudness for its own sake.
Learn while you work
The Learn section explains the ideas behind the tools in plain language: what LUFS means, how a mastering chain is ordered, how to mix a vocal, and how to make a female version cover from a separated reference.
Featured releases
All releases
Latest guides
All guidesTeentaal: the 16-beat cycle of Hindustani music
Teentaal is a 16-beat Hindustani rhythmic cycle counted in four groups of four, with claps on 1, 5 and 13, and a wave on 9. Its theka, or basic tabla stroke pattern, is dha dhin dhin dha | dha dhin dhin dha | dha tin tin ta | ta dhin dhin dha.
Taal for beginners: teentaal, keherwa and dadra with clap patterns
A taal is a repeating rhythmic cycle: count 16 beats for teentaal, eight for keherwa, and six for dadra. For teentaal, clap on 1, 5 and 13, wave on 9; for keherwa, clap on 1, wave on 5; for dadra, clap on 1, wave on 4.
Rupak: the 7-beat taal that starts on khali
Rupak is a taal, or repeating rhythmic cycle, with seven beats grouped 3 + 2 + 2. Its opening closed strokes make sam, the first beat, sound light; this beat is khali, marked by a wave, rather than tali, marked by a clap.
Keherwa: the 8-beat taal of film songs, bhajans and folk
Keherwa is an eight-beat taal, or repeating rhythmic cycle, arranged as 4 + 4. To recognise it, listen for two groups of four, a lighter second half, and the recurring drum pattern: dha ge na ti | na ka dhin na.
Jhaptaal: the 10-beat taal
Jhaptaal is a taal, a repeating rhythmic cycle, with ten beats. Its four vibhags, or divisions, follow 2 + 3 + 2 + 3, alternating two-beat and three-beat groups.
Ektaal: the 12-beat taal of khayal
Ektaal is a twelve-beat taal, or repeating rhythmic cycle, divided into six two-beat sections called vibhags. Its stretched beats let the tabla player elaborate the stroke pattern in vilambit (slow-tempo) bada khayal, the slow opening composition of a khayal performance.
Questions people ask
Are the BeatRootRaga audio tools really free?
Yes. Mastering a track and separating a song into stems are free, with no subscription. An account is only needed to keep results in your dashboard and to run full-length stem separations.
Do I need to install anything to use the BeatRootRaga tools?
No. Both tools run in the browser on desktop and mobile. You upload a file, preview the result and download the output.
Which audio formats do the BeatRootRaga tools accept?
MP3, WAV, FLAC and AIFF. For mastering, upload the final mix with a few decibels of headroom rather than an already limited file.
What loudness should I master to for Spotify or Apple Music?
Around minus 14 LUFS integrated with true peaks at or below minus 1 dBTP suits streaming platforms, which turn louder tracks down. The mastering tool lets you set the target and preview it.
Can I use the stems commercially?
Separating a song does not change who owns the recording. Use stems only from music you own, have permission to use, or that is licensed for your purpose.
Which AI model does the BeatRootRaga stem splitter use?
Demucs, an open-source hybrid transformer model for music source separation. The preview and the full song use the same model.