Capturing Audio & Screen
Capture your computer’s/meeting audio and record your screen, all stored locally.
The Kalima desktop app can capture the audio coming out of your computer, the meeting itself, and optionally record your screen alongside it. Everyone on the call gets transcribed, not just your microphone, and any video you record stays on your own machine.
Screen and video recordings are saved as local files on your computer and are not uploaded to the cloud. They stay under your control on your own device.
Capture audio or screen
Use the tabs below to set up the capture you need. Both share the same account and balance as the rest of Kalima. Capturing meeting audio, transcription, and translation draw from your normal plan usage.
Capture your computer's / meeting audio
The desktop app records the sound playing on your computer so remote participants on a call are transcribed too. You can mix in your microphone so both sides of the conversation land in one transcript.
When you record from a detected meeting, Kalima automatically captures both your microphone and the meeting (desktop) audio, with your mic included by default.
Native audio source picker (Windows)
On Windows, Kalima can isolate the audio of a single running app so other system sounds don't bleed into your transcript. When you start an isolated capture, a picker appears with thumbnail previews to help you choose the right source.
If the isolated app audio is silent or unavailable, Kalima automatically falls back to capturing your whole system's audio, and then to your microphone, so a recording never ends up empty.
Targeting a single app keeps notification sounds, music, and other tabs out of your transcript. On macOS and Linux, Kalima captures system-wide audio through your operating system's screen/audio sharing permission instead.
What's available where
A few of the most advanced capabilities rely on Windows. On macOS and Linux you still get system-audio capture and screen recording through your operating system's sharing permission.
| Capability | Windows | macOS / Linux |
|---|---|---|
| Capture meeting / system audio | Yes | Yes (via OS sharing permission) |
| Mix in your microphone | Yes | Yes |
| Isolate a single app's audio | Yes | System-wide audio |
| App / window source picker with previews | Yes | No |
| Screen / video recording (saved locally) | Yes (browser-free) | Yes |
| Crash recovery of partial recordings | Yes | Yes |
Screen recording requires a Plus plan or higher. On the Free plan the screen-recording option is locked and shows an upgrade prompt. Capturing meeting/system audio is available on every plan.
Frequently asked questions
Yes. The desktop app captures your computer's meeting audio and can mix in your microphone, so both sides of the conversation are transcribed. On Windows you can isolate a specific app's audio; on macOS and Linux, Kalima captures system audio (you may be asked to grant a screen/audio sharing permission).
Your recording is probably capturing the microphone only. Start the recording with a capture mode that includes meeting/desktop audio, and grant any screen/audio sharing permission your operating system requests. On Windows, you can pick the exact app to capture in the source picker.
Screen and video recordings are saved as local files in a recordings folder on your computer (your Documents folder by default), which you can change in settings. You can list them, open the folder, delete them, and play them back inside Kalima. They are not uploaded to the cloud.
No. Screen and video recordings stay local on your machine and are not synced to the cloud, so they don't count against your plan's cloud storage. Your transcript itself is saved to Kalima as usual.
Isolating one app's audio is a Windows-only capability. On macOS and Linux, Kalima captures system-wide audio through your operating system's screen/audio sharing permission instead. The same applies to browser-free native screen recording, which is also Windows-only.
If isolated app audio is silent or can't be captured, Kalima automatically falls back to capturing your whole system's audio, and then to your microphone, so the recording still works.