Plenty of songs start as a voice note: a melody hummed into the phone, a bhajan sung at home, a verse recorded after band practice. Turning that into a full song used to need a producer, or at least hours in a music app. With AI you can build the band around your voice yourself, one instrument at a time, and keep control of what goes in.
On Autunes this is Layer Studio. Unlike tools that turn your vocal into a finished track in one go, it adds parts one by one, each on its own lane, so you can hear what every instrument adds and take out anything that crowds your voice.
Step 1: Record a clean vocal
The better the vocal, the better the band fits it. You do not need a studio, but a few habits help.
- Record in a quiet, soft room: a bedroom with curtains and a bed beats a tiled bathroom.
- Keep the phone at the same distance the whole time, about a hand's width from your mouth.
- Sing to a steady pulse. Tap your foot or play a metronome in headphones so the timing stays even.
- Leave a second or two of silence at the start and end.
Step 2: Upload it to Studio
In Studio, choose Upload a track. It takes MP3, WAV, FLAC, M4A or OGG up to 50 MB and 10 minutes. Choose the language you sing in so the words are transcribed, and tick the box confirming the recording is yours. The upload becomes a normal track in your library.
Step 3: Add the first layer
Open the track and choose Layers, then Start layering. Pick an instrument from the twelve types: Drums, Percussion, Bass, Guitar, Keys, Strings, Winds, Brass, Synth, FX, Lead vocal or Backing vocals. Describe the part in plain words and press Add part.
Start with whatever holds time and harmony together. For most pop and film-style songs that is drums or percussion, then bass, then a chord instrument like guitar or keys. For a bhajan it might be tabla and harmonium-style keys first.
- Percussion: "soft tabla theka, simple in the verse, fuller in the chorus"
- Bass: "warm bass following the kick, simple, leaves space for the voice"
- Guitar: "gentle fingerpicked acoustic guitar, arpeggios"
- Strings: "slow string pad only in the chorus, cinematic"
- Winds: "bansuri melody answering the vocal between lines"
Step 4: Build up, one part at a time
Each new part is written while the AI hears your vocal and every lane already in the project, so the bass locks to the drums and the strings sit under the voice. To put a part in only one section, drag on the ruler first; the next part then covers just that stretch, such as a shehnai only in the intro.
Use mute and solo to judge each layer. If the guitar fights the vocal, mute it and try another description. Changing volumes, muting and soloing costs nothing.
Step 5: Export
When the balance is right, press Export. Autunes flattens your vocal and every unmuted lane into one WAV, at the volumes you set, ready to upload or to keep working on in another app.
Add part vs One-shot
If you only want to hear an idea quickly, One-shot remixes the whole song with the new instrument in it and saves a new track. It is faster but gives no lanes to mix. Add part is slower and keeps every decision yours.