LASTUDIO
Blog
Guides

How to Convert Humming to MIDI for Free — Right in Your Browser

What People Really Want to Know About Humming to MIDI

"I've got a melody in my head but I can't read or write music." "I can't play piano, but I want to get my idea into a DAW as actual notes." — This is exactly where voice-to-MIDI conversion comes in. In this guide, we'll cover everything: which free, browser-based tools to use, how to get the best conversion results, and what to do with your MIDI once you have it.

The short answer: as of 2025, you can convert a hummed melody to MIDI entirely in your browser. No software to install, no high-end PC required. Just hit record, hum your melody, and within seconds you'll have an editable MIDI track.

Musician humming and composing in a studio

How Does Humming-to-MIDI Actually Work?

Converting voice to MIDI used to require expensive pitch-detection software and a fair amount of technical know-how. That changed dramatically with the rise of machine learning–based pitch estimation.

The Three-Step Process

At a high level, voice-to-MIDI conversion works in three stages:

  1. Audio Input (microphone or file upload): Your voice is captured as a digital audio signal.
  2. Pitch Estimation (frequency → note name): The audio waveform is analyzed to detect the fundamental frequency (F0) at each moment, which is then mapped to the nearest musical note (C4, D4, etc.).
  3. Note Boundary Detection (finding where notes start and stop): The tool identifies the onset and offset of each note and encodes them as MIDI note-on and note-off events.

The Key Technologies Behind It

  • Basic Pitch (by Spotify): Runs an ONNX model directly in the browser and can generate MIDI even from polyphonic audio. Spotify's open-source release made this widely accessible.
  • CREPE / pYIN: Classical algorithms optimized for monophonic pitch tracking — great for humming or whistling single melodies.
  • WORLD / HarvestF0: F0 estimation algorithms developed in the speech synthesis research community, known for high accuracy and low latency.

All of these share one key principle: the cleaner and more single-pitched your voice, the better the result. Mixed voices or chords will confuse the detector, so aim for one clear note at a time.

Free Voice-to-MIDI Tools Compared

Several browser-based tools can handle voice-to-MIDI conversion. Here's how they stack up.

1. LA Studio (Browser DAW) — Built-in Voice to MIDI

LA Studio is a completely free, browser-based DAW with a built-in Voice to MIDI feature. Record your hummed melody and it automatically converts to a MIDI track. You can then edit notes directly in the piano roll, audition them with built-in synths or SoundFonts, and export — all without leaving the browser. No account, no install, no friction.

2. Basic Pitch (by Spotify) — File Upload

Basic Pitch lets you upload an audio file and download the resulting MIDI. Its polyphonic capability is impressive, though it's less relevant for humming since you're singing one note at a time. The main drawback: you have to bring the exported MIDI into a separate DAW to do anything with it, which adds steps to your workflow.

3. Melodyne (Paid Desktop App) — Professional Grade

Celemony Melodyne is the industry standard for pitch correction and MIDI export. The accuracy is top-tier, but even the entry-level Essential edition costs around $100. A 30-day free trial is available, but it's overkill if you just want to quickly capture a melody idea.

4. GarageBand (Mac/iOS) — Free but Apple-Only

GarageBand can convert a vocal recording into a MIDI region, and it's free on Apple devices. That said, it's not an option on Windows or Chromebook.

Bottom line: if your requirement is "free, right now, on any OS, no install", LA Studio's Voice to MIDI feature is the most practical choice.

Musician recording into a microphone

Step-by-Step: How to Convert Your Humming to MIDI

Here's a walkthrough using LA Studio's Voice to MIDI feature.

Step 1: Set Up Your Microphone

  1. Use your computer's built-in mic, a USB mic, or a headset mic — any of these will work.
  2. Find a quiet space. HVAC noise, fans, and ambient sound all degrade conversion accuracy.
  3. Position the mic about 6–8 inches from your mouth. Too close causes plosives; too far reduces signal clarity.

Step 2: Open the Voice to MIDI Panel in LA Studio

  1. Open https://la-studio.cc/editor in your browser (Chrome recommended).
  2. Once the editor loads, select the "Voice to MIDI" option from the track creation area.
  3. When the browser asks for microphone access, click "Allow."

Step 3: Record Your Melody

  1. Click the red record button to start.
  2. Hum or sing your melody using a vowel sound like "laaaah" or "naaaah." Slightly slower than normal is better.
  3. Break it into short phrases — 4 to 8 bars at a time — for cleaner results.
  4. Click stop when you're done.

Step 4: Review and Edit the MIDI

  1. After you stop recording, pitch analysis runs automatically — usually just a few seconds.
  2. The detected MIDI notes appear in the piano roll.
  3. Play it back and drag any off notes into place using the piano roll editor.
  4. Assign an instrument (piano, strings, synth, etc.) to hear your melody with a real sound.

Step 5: Export as a MIDI File

  1. Go to the export menu and choose "MIDI file (.mid)."
  2. The exported file can be imported into any DAW — Cubase, Logic Pro, Ableton Live, FL Studio, and others.

7 Tips for Getting Better Conversion Results

Voice-to-MIDI isn't perfect, but these habits will cut down your editing time significantly.

1. Sing One Clean Note at a Time

Heavy vibrato and ornaments cause the pitch detector to create lots of tiny fragmented notes. Start by practicing sustaining a single pitch steadily before hitting record.

2. Keep a Steady Tempo

Singing along to a metronome or a click track makes note lengths come out much more consistent. Tempo fluctuations translate directly into rhythmic inaccuracies in the MIDI.

3. Stay in a Comfortable Range

Pitch detection is most reliable in the mid-range. For most people, roughly C3–C5 (men) or C4–C6 (women) will give you the best accuracy. Extreme highs and lows tend to be less reliable.

4. Minimize Background Noise

Air conditioners, computer fans, and street noise all interfere with pitch detection. Check your environment before recording, and consider using AI noise removal to clean up the audio afterward.

5. Keep Takes Short

Try to keep each recording under 30 seconds for stable processing. Record longer melodies in two or three segments, then stitch the MIDI together in the piano roll.

6. Set the Right Recording Level

Both too quiet and too loud will hurt conversion accuracy. Do a quick test recording first and adjust your gain so the level meter sits around −12 dB to −6 dB.

7. Always Clean Up in the Piano Roll

No tool is perfect — you'll almost always get a few notes that are a half-step off or very short ghost notes that shouldn't be there. Budget 10–15 minutes of piano roll editing after every conversion to bring it up to final quality.

What to Do With Your MIDI Once You Have It

A MIDI melody captured from humming is already useful as-is, but here are some ways to take it further.

Analyze the Harmony and Add Chords

Import the melody MIDI into your DAW, detect the key, and use an auto-accompaniment plugin or chord suggestion feature to build out an arrangement quickly. LA Studio's BPM and key detection tool is handy for this step.

Print or Share as Sheet Music

Import the MIDI file into MuseScore (free) and it instantly generates readable notation. This is a great way to transcribe a melody you picked up by ear.

Use It as a Seed for AI Music Generation

Render your MIDI as audio and feed it into an AI music generation tool as a reference. You can get automatic orchestral arrangements or jazz harmonizations built around your original idea.

Trigger Software Synths

Assign a synth like Vital, Dexed, or Surge XT to your MIDI track in LA Studio and hear your hummed melody played back on a lush synth lead or grand piano. It's a great motivational boost when you're writing.

DAW piano roll showing MIDI editing

Voice to MIDI vs. Audio to MIDI — What's the Difference?

These two terms get confused, so here's a quick breakdown:

  • Voice to MIDI: Converts a sung or hummed vocal performance — live or recorded — into MIDI. Designed for monophonic (single-pitch) input.
  • Audio to MIDI: Converts any audio file — a full song, a guitar riff, a piano recording — into MIDI. Some tools (like Basic Pitch) support polyphonic audio.

For capturing a melody you're humming, Voice to MIDI is the right tool. For converting an existing guitar or piano recording you have on file, Audio to MIDI (Basic Pitch) is a better fit. LA Studio supports both, so you don't need to switch tools.

Wrap-Up: Hum a Melody, Get MIDI — Free, in Your Browser, Right Now

Converting a hummed melody to MIDI used to mean expensive software and a steep learning curve. In 2025, it takes about 30 seconds in a browser.

  • ✅ No installation or account required
  • ✅ Works on Windows, Mac, and Chromebook
  • ✅ Edit notes immediately in the piano roll after conversion
  • ✅ Export standard .mid files compatible with any major DAW

If you've ever lost a melody because you had no way to capture it, or if you want to start composing without being able to play an instrument, give LA Studio's Voice to MIDI feature a try. Hit record, hum your idea, and watch it appear as notes on the screen — it might just change how you think about songwriting.

Frequently Asked Questions

Q. Do I need any music theory knowledge to use this?

A. Not at all. The tool handles pitch analysis and note generation automatically. You don't need to know note names or how to read music. After conversion, you can adjust notes by ear directly in the piano roll.

Q. How accurate is the conversion? Will it get notes wrong?

A. For a clear, steady hum in a comfortable vocal range, expect around 80–90% of notes to be detected correctly. Vibrato, ornaments, and unstable tempo will reduce that number. Always review and clean up the result in the piano roll.

Q. Can I use this on my phone?

A. The browser-based tool works best on a desktop or laptop (Chrome recommended). It may run on a mobile browser, but WebAudio API limitations can cause performance issues on phones. We recommend using a PC for this workflow.

Q. Can I import the exported MIDI into Ableton Live or Cubase?

A. Yes. The exported .mid file is in the standard MIDI format, which is compatible with virtually every major DAW — Ableton Live, Cubase, FL Studio, Logic Pro, Studio One, Reaper, and more. Just drag and drop it into your project.

Q. What if I try to hum chords — will that work?

A. Voice to MIDI is designed for monophonic input (one pitch at a time), which is essentially all a human voice can produce anyway. So in practice, this isn't a limitation. If you need to convert a polyphonic audio file — like a guitar chord progression — use Basic Pitch (Audio to MIDI) instead.

Related Articles

News
The Complete Suno AI Guide: 5 Prompting Tips That Actually Work [2025]
Master Suno AI prompting from scratch — covering genre stacking, instrument keywords, metatags, and how to bring your generated tracks into a DAW.
Reviews
ブラウザDAW クラウド保存&共有機能を徹底比較【2026年版】
無料ブラウザDAWのクラウド保存・共有・コラボ機能を徹底比較。選び方のポイントも解説。
Guides
How to Convert SF2 to SFZ for Free [Polyphone & Browser DAW]
A complete guide to converting SF2 soundfonts to SFZ using the free tool Polyphone, plus how to use them directly in a browser DAW.