Audio devices

Settings › Audio input / output…, or the sample-rate readout beside the project name, chooses the recording input and the playback output for the whole studio. Both workspaces share the choice: the input feeds the single-file recorder, multitrack recording and live monitoring, and the output carries everything you hear. Input and output are separate decisions, so you can record through an interface while listening on headphones plugged into the computer, or the other way round.

When to use it

Open it when you plug in an interface, when a take comes out of the wrong microphone, or when the sound should go to a different output than the system’s default. The panel lists devices as the browser names them and refreshes when you plug something in while it is open.

Controls

Recording input

System default input follows the browser’s and the system’s choice. Choosing a device by name asks for exactly that device: if it is missing later, recording stops with a message pointing back here rather than quietly falling back to the laptop microphone. Changing the input while monitoring reconnects the monitor to the new device.

Allow access / refresh devices

Browsers only reveal device names after you have granted microphone permission. This button asks once and lists the devices; the stream it opens is released immediately. Opening the panel alone never requests access.

Playback output

Where the browser supports choosing an output, the list switches every running part of the app to the chosen device, including the sound library demos. Some browsers add their own output chooser instead; browsers without output selection show a note to change the system output. A failed switch keeps the previous device and setting.

Buffer size

How far ahead the browser prepares audio. Low, Balanced (the default), High and Max trade responsiveness for safety: a larger buffer rides out a heavy project or a busy computer without crackling, but everything you hear live arrives later, including monitoring, live MIDI and key previews. As a guide, an interface’s output adds about 8 ms at Low, 16 ms at Balanced, 27 ms at High and 90 ms at Max. Choose Low while playing or singing along, and High when a dense mix crackles. Changing it restarts audio; playback continues from the same place.

Recording latency

A recorded take arrives a little late: the sound you heard left the output, and your playing came back through the input. Dr. Tape moves every audio take earlier by that round trip, so it lands where you played it. MIDI takes move earlier by the output part alone.

Until you measure, the round trip is estimated from the browser’s figures, which can be several milliseconds off, more at High and Max. Measure with a loopback cable… makes it exact: connect a cable from the left output (Output 1) of your interface to the recording input, turn off phantom power and the interface’s direct monitoring, and keep the volume low, then start. Short noise bursts play on the left channel, three times at every buffer size, about 45 seconds in total, and the table shows what was measured next to the estimate. Even measured, a take can land up to about 3 ms off at Low, because the browser’s input timing shifts slightly each time audio starts. The result belongs to those two devices and the project’s sample rate; measure again after changing either.

During a take

The selectors are disabled while recording or while another operation is busy, so a take cannot change device half way.

Tips

• The choice is remembered per browser and is not part of the project; a project opened on another machine uses that machine’s devices.
• If a saved device shows as unavailable or needing permission, pick it again after reconnecting; nothing switches without your choice.
• This picks a device, not a channel: capture is mono, so choose the input channel on the interface itself.
• The buffer size and latency measurements are remembered per browser, like the devices.