Hummed composition sketches
Capture a tune before it is arranged, then assign the downloaded notes to any instrument in your DAW.
Voice to MIDI Converter
Choose a recording of your singing or humming and turn it into editable MIDI on this device. Listen to the detected melody before downloading it for your music software.
Sing separate “da” or “la” syllables with clear attacks. A continuous “oooh” glides between notes and is more likely to fragment.
6.5-second synthetic melody · a workflow demo, not an accuracy benchmark.
This file may need a lot of memory
The selected file is {size}. MIDIFLOW does not upload it or impose a server limit, but very large files can run out of memory, especially on a phone.
0notes
—range
0.0sduration
0max polyphony
0.25
Compare the original with a simple synth preview of the detected notes. The preview does not reproduce the original sound or pitch bends.
Too many or missing notes? — increase it to keep fewer, more confident notes; decrease it to include quieter notes. Then compare the previews again.
Free foreverNo signupNo watermarkAudio never uploaded
Best input
Why a single voice can transcribe well
A solo vocal usually presents one intended pitch at a time. When note attacks are deliberate, the model can follow a melody without separating chords or instruments.
Clear sung syllables
Repeated consonant-vowel syllables create visible onsets and help separate one note from the next.
Comfortable solo range
A steady chest or mixed voice recorded close to the microphone offers a stronger fundamental than breathy extremes.
What can go wrong
Vocal gestures that split or bend MIDI notes
Human voices use portamento: pitch often slides into a target rather than jumping instantly. A long transition can become a bend or several adjacent notes depending on its speed and stability.
Breathy singing adds broadband noise while falsetto can have a weak fundamental. The model may follow a stronger harmonic, drop a quiet note or report an octave shift.
Unaccompanied singers also drift in pitch and timing. A held “oooh” between note targets offers no consonant onset, so note boundaries are less obvious than they sound in your head.
Convert audio to MIDI in three steps
- 01
Choose audio
Choose an MP3, WAV, M4A, FLAC or OGG recording, or try the sample. Start with a short solo part to check the result.
- 02
Transcribe notes
The converter estimates notes on this device. Your recording is not uploaded and no account is needed.
- 03
Listen, adjust and download
Compare the original with Play MIDI notes, apply a different note threshold if needed, then save the .mid file for your music software.
Transcription is powered by Spotify's open-source Basic Pitch model and runs in your browser.
Improve accuracy
Sing a transcription-friendly melody
Perform for clear note events, not for a finished vocal sound. You can add legato, expression and human timing after the melody is in MIDI.
Use “da” or “la” on every note
The consonant creates a small attack and the vowel carries pitch. Avoid one continuous “oooh” across the whole phrase.
Follow a metronome
A click in headphones helps separate intended note lengths and makes the resulting MIDI easier to quantize without changing the melody.
Stay in a comfortable range
Choose a key where you can sing steadily without forcing low notes or switching into unstable breathy falsetto.
Import the MIDI into your DAW
The download is a standard MIDI file containing note timing, pitch, velocity and eligible pitch-bend events. It does not contain the original instrument sound, so assign any software instrument after importing it.
Ableton Live
Create or choose a MIDI track, then drag the .mid file into an empty clip slot or the Arrangement. Load an instrument on that track and edit notes in the MIDI Note Editor.
Logic Pro
Drag the file into the Tracks area and choose a Software Instrument track. Open the Piano Roll to correct timing, note lengths or velocity before arranging.
FL Studio
Drag the file into the Channel Rack or use File › Import › MIDI file. Send the imported notes to a chosen instrument, then clean them in the Piano roll.
FAQ
Can I record my voice directly on this page?
This page converts existing audio files. Record your melody in your phone or computer recording app, save the file, then select it here. MIDIFLOW does not currently offer live microphone recording or real-time MIDI output.
Is humming easier to convert than a full vocal recording?
A solo humming or singing track is usually easier than a produced vocal mixed with instruments. Clear syllable attacks are more important than using lyrics.
Why does one sung note become several MIDI notes?
Pitch drift, vibrato and portamento move through several pitch regions. The model may segment that continuous motion, especially when the onset is unclear.
Should the metronome be audible in the recording?
No. Listen through headphones so the click guides your timing without entering the microphone and creating extra transients.
Can I sing lyrics instead of da or la?
Yes, but changing consonants, breath and vowel shapes make the signal less consistent. Repeated da or la syllables are a cleaner way to capture a compositional sketch.
Will the MIDI preserve my vocal timbre and words?
No. MIDI contains detected notes and performance data, not lyrics or the sound of your voice. Choose a software instrument after importing it.
Can two singers be converted at the same time?
The model is polyphonic, but overlapping voices share similar timbres and move independently, so a duet is much less reliable than recording each vocal line separately.