Headphone-free recording
Now, a reality.
Songzap V3 introduces ground-breaking AI technology that makes headphone-free recording a reality. Yes, you read that right! You can now enjoy multitrack recording without having to use headphones. Our pioneering HFR machine-learning technology isolates playback ‘bleed’ from backing tracks and overdubs, freeing you from having to wear headphones or use any kind of external device. Just sing or play straight into your iPhone or iPad mic, record layers and get ready to be stunned. Crystal-clear arrangements with no artefacts or spill. What’s more, it works in real-time, meaning the bleed removal is instantaneous, with no need to wait for any online processing to run or complete. Check out the videos below to see two of our exciting artists giving Songzap V3 and HFR a spin.
Our mission statement has always been to boost creativity for you, the music makers. We do so by removing technological barriers through intuitive and innovative designs. While developing V3, the Songzap team collaborated with a large pool of creative musicians, working in different musical styles and contrasting workflows. What we learned was that music makers want zero barriers when inspiration strikes. Even the notion of having to plug in headphones (or look for adaptors) when performing into your mobile device had to go. The result?
Got a lyric in your head while you’re commuting or exploring the outdoors? A top line for that backing track you’ve been carrying around? Rhymes for that loop stuck in your head? Some harmonies while you’re at it? Get your phone out, capture your ideas, layer them and mix it all before you even hop on that train. Before the moment goes.
You won’t even notice that a clean-up process is taking place. HFR works while you record. By the time you hit ‘stop’, your new recording has already been isolated from any backing tracks or previous recordings blasting out of your phone or tablet speaker. Unlike current stem-isolation technologies, Songzap’s Headphone-Free Recording algorithm not only works in real-time but it is also source agnostic. That means, it will isolate elements such as vocal recordings from previously captured vocal layers – perfect for songwriters and top-liners creating ideas on the go.
HFR works both in ‘Tracking’ and ‘Loop Record’ modes with a Songzap Premium subscription. A neat HFR button in either mode brings up a fader that lets you check ‘before/after’ results; or blend in a bit of ‘spill’ if you want it.
Before you can make headphone-free recordings, Songzap will first need to calculate the audio latency of your device. This is so that the AI can accurately remove any playback bleed from your recordings. When you first try to make a headphone-free recording, Songzap will run the latency calibration routine: a short audible sound will be made and the app will identify everything it needs to enable headphone-free functionality. To run the latency calibration routine, you’ll need to make sure that your device volume is up, you are in a quiet area with minimal background noise, and no headphones or audio interface are connected.
Check out the videos below for an inspired vocal-layering demo and a quick Headphone Free Recording tutorial using Songzap V3.
Endless fun and layering creativity unlocked.
Headphone-free recording
vs. Conventional stem separation.
| Feature | Headphone-Free Recording | Conventional Stem Separation |
| Purpose | To prevent / remove bleed in real-time, so that recordings can be made without headphones being worn. | To split mixed audio into component (instrument) parts for remixing, processing, or analysis. |
| Use cases | Songwriting, demoing, layering vocals without headphones, on-the-go mobile recording. | Remixing, mastering, stem-level mixing, sampling, reconstruction of multitrack material. |
| Source dependency | Source agnostic – removes a live recording from playback bleed, regardless of source type. (It can therefore isolate a live vocal from a previously recorded vocal.) | Source dependent – removes a specific instrument type from a mix of different instrument types. (Cannot separate vocals from vocals or guitar from guitar.) |
| Speed of operation | Processes in real-time while a recording is being made and captured. On completion of the recording, the isolation is also complete. | Offline process. A previously recorded audio files is stem-separated (often in the cloud), as a post-production process. Longer files take longer to process. Not suitable for real-time or live performance/recording use. |
| User interface | An audio waveform which shows sound that is both in the foreground (i.e. the recording) and background (i.e. playback bleed). A user control allows the amount of bleed isolation to be manipulated. | An audio waveform is processed/split into multiple waveforms designated as instrument types, such as voice, guitar, bass, piano, drums. |