Opidocs
FeaturesVoice Capture

Troubleshooting

Fix common issues with Voice Capture — microphone permissions, browser support, and recording errors.

Quick fixes for the most common Voice Capture issues.

Microphone permission denied

Your browser needs permission to access your microphone before Voice Capture can work. If you accidentally denied access or need to re-enable it:

  1. Click the lock icon (or site settings icon) in your browser's address bar
  2. Find the Microphone setting and change it to Allow
  3. Refresh the page and try recording again

On mobile, you may also need to check your device's app permissions in system settings.

No transcript appearing

If you see "Listening..." but no words appear on screen:

  • Check your microphone — Make sure the correct microphone is selected in your system sound settings and that it's picking up audio
  • Speak at a normal volume — Very quiet speech may not be picked up clearly
  • Try a different microphone — If you have multiple microphones connected (built-in, headset, external), switch to another one
  • Check your internet connection — Transcription requires a live connection to the server. A slow or dropped connection will prevent text from appearing

Recording stops unexpectedly

If your recording ends without you clicking Stop, the connection to the transcription service may have dropped. Opisense reconnects automatically up to 3 times. If all reconnection attempts fail:

  • Your audio backup is preserved — you won't lose your recording
  • Try starting a new recording session
  • Check that your internet connection is stable

"Active recording" error

Only one recording session can run at a time across all of Opisense. If you see an error about an active recording:

  • Check whether you have a recording running on the Voice Capture page
  • Check other browser tabs for open documents with an active transcript block
  • Stop the existing recording, then try again

No speaker labels appearing

If you're recording a meeting or conversation but don't see speaker labels in the transcript:

  • Speaker detection is automatic — it only shows labels when the transcription service identifies more than one distinct speaker. If everyone sounds similar or only one person is speaking, labels won't appear.
  • Speak one at a time — Overlapping speech makes it harder to distinguish speakers. Taking turns produces better speaker identification.
  • Use the right environment preset — In noisy environments, try switching to the Office or Noisy preset. Background noise can interfere with speaker detection.
  • Microphone placement matters — A single laptop microphone far from participants may not pick up enough difference between voices. If possible, position the microphone centrally or use a conference microphone.

Speaker detection works best with 2-4 distinct speakers. The quality depends on the audio the transcription service receives — clear audio with distinct voices produces the best results.

Recording too short

Recordings under 2 seconds are discarded automatically. This prevents accidental taps from creating empty documents. If your recording was intentionally short, try speaking for at least a few seconds before stopping.

Browser compatibility

Voice Capture requires a modern browser with WebRTC and MediaRecorder support:

BrowserSupported
ChromeYes
EdgeYes
FirefoxYes
Safari14.1+

Mobile browsers are supported but may have limited microphone access depending on your device's OS permissions.

If you experience a crash or unexpected page close during recording, Opisense keeps a backup of your audio. A recovery dialog will appear the next time you open Voice Capture, giving you the option to restore your recording.

Having trouble on mobile? See the mobile Voice Capture troubleshooting section for issues specific to the mobile app — including microphone permissions, background recording, and crash recovery.

On this page