macOS app: no transcription progress indicator after dictation stops

Open 💬 1 comment Opened Jul 18, 2026 by hacksurvivor

What version of the Codex App are you using (From “About Codex” dialog)?

26.715.31925

What subscription do you have?

Paid ChatGPT subscription (exact tier not included in this report)

What platform is your computer?

Darwin 25.5.0 arm64 arm

What issue are you seeing?

When using dictation in the Codex app, audio recording starts and the timer/stop control is visible during capture. After stopping the recording, however, there is no loading spinner, “Transcribing…” status, or other progress feedback while the audio is being processed. The composer looks idle, so it is impossible to tell whether the speech was captured, transcription is still running, or the operation has stalled. The same regression is also visible in the ChatGPT macOS app.

What steps can reproduce the bug?

  1. Open the message composer in the Codex macOS app.
  2. Start dictation and speak a message.
  3. Stop the recording.
  4. Observe the composer while waiting for the transcript.
  5. There is no spinner or transcription-in-progress state until text eventually appears (or the operation fails).

The same missing feedback occurs in the ChatGPT macOS app.

What is the expected behavior?

Immediately after recording stops, the composer should show a persistent and accessible transcription state, such as a spinner with “Transcribing…”, until the transcript is inserted or an error is shown. On failure, the app should provide a clear retry action and should not silently discard the capture.

Additional information

This began after recent app updates. The transcription may still complete; the bug is the missing in-progress feedback, which makes the app appear unresponsive and leaves the user unsure whether their speech was captured. Screenshot attached showing the dictation composer controls.

<img width="726" height="42" alt="Image" src="https://github.com/user-attachments/assets/81a34a4e-1c52-4e80-9e4a-f50b9d1f5f45" />

View original on GitHub ↗

This issue has 1 comment on GitHub. Read the full discussion on GitHub ↗