When I'm using double-tap to start/stop recordings, I have zero way to know if Omi is actually recording without digging out my phone. If it's in my pocket and I'm in the middle of an important conversation, that means unlock phone → open Omi app → pray Bluetooth is connected → check status. That's a lot of friction for something that should be instant. What I'm asking for is simple: give me a physical indicator on the Omi device. 🔴 LED light — solid or pulsing when recording is active. Red, green, whatever. Just something I can see at a glance. ⏰ Haptic/buzz pattern — a distinct buzz on start and another on stop so I can feel the confirmation on omi without looking at my omi or device. In high-value conversations — client meetings, important calls, something I know I'm going to want to remember — I need to know I'm capturing it in real time, not wonder about it afterward. Every wearable and recording device worth its salt has some version of this. It's table stakes UX. This would be a big productivity win for anyone using Omi in professional settings.
Hello, I want to connect the Google calender extensions app and use more agenda’s that I have on Google calender. When I log in via the extension, I see “the app is blocked” with explanation. If I use the integration on OMI, I only get the “main” agenda. What can I do to fix both?
CRITICAL BUG 3.0.21 - No Voice, No Transcription, No Photos After updating to firmware 3.0.21 my Omi CV1 stopped working completely. Device: Omi CV1 Device ID: 839FE8E1-4F16-013F-26FA-6E17F772F076 Email: arbat77@icloud.com iOS Issues: No voice / 2-way conversation broken No transcription in any mode Photos not syncing I tested all 4 transcription modes: Omi, On Device, Cloud, Parakeet. None of them work. Screenshot attached. Error from logs: [OnDeviceApple] Transcription error: PlatformException(RECOGNITION_ERROR, Failed to access assets, null, null) Please fix ASAP! Screenshot of logs attached
Connectivity is up and down. All day it connects and disconnects. I am wearing it as a watch, but it constantly drops and picks up. Also the device runs out of power before my evening meetings. So I power it every afternoon. It should last all day!
When I attempt to connect with gmail, it is saying 3 things are already connected (calendar) but wouldn’t let me add gmail even though the app opened.
I'm not sure why, everything was fine 2 days ago, but today (was not used yesterday) after 6 hours of “listening” with the app, none of the recordings/transcriptions/data was saved. It didn't sync later, and even the Omi AI said it was probably lost. Not sure if it's to do with the recent update or what. But figured I'd at least mention it in case it's not just user error lol
The battery on my Omi isn't lasting as long as it did a few weeks ago. The last few days it's only lasted about 10 hours until it dies. I turned on the VAD gate, too.
One of my favorite features is missing. I recently installed the beta version of this app, and it acted strangely. I have to return to the current version, but the graph didn’t load for me. Can someone please fix it? Also, could we animate it? It would be awesome if we could animate this like the human nervous system that fires up with electricity.
Device: Omi CV1 (consumer pendant), firmware 3.0.21 App: Omi iOS, latest App Store version (Sept 22, 2026) Primary language: Chinese (Mandarin, Traditional) Network: tested on both Wi-Fi and 4G Description: The "Teach Omi your voice" / speech profile step never completes. Every attempt stays at 0% on the progress bar and then shows a dialog "Connection lost — please check your internet connection and retry". Tapping Retry produces the same result immediately. What I tried (all failed the same way): Wi-Fi and 4G, VPN off. Primary language switched to English and Auto-translate turned off, speaking English continuously for 30+ seconds. Force-closing and reopening the app between attempts. Device is connected (blue light), battery 70%+, live transcription on the home screen works fine at the same time. Everything else works: pairing, live transcription in Chinese, speaker 1/2 diarization, summaries. Only speech profile enrollment fails, so transcripts never label me as "You". Related: I also tried tagging my own segments and another person's segments in finished conversations. Labels are saved, but the next conversation starts again as Speaker 1 / Speaker 2 with no suggestion — no voiceprint seems to be created. This matches GitHub issue #7187 (profiles only work for English speakers) and #12765. Because I attempted enrollment many times, there may now be stale/partial speech samples on my account (see GitHub issue #5266). If so, please clear them — there is no reset option in the app. Expected: Progress bar advances while speaking; profile is stored; later conversations identify me as "You", regardless of transcription language. Actual: 0% → "Connection lost" on every attempt, in both Chinese and English. Happy to provide account email, exact timestamps of attempts, or a screen recording.
Hello! I can’t switch from Cloud to Local mode because the button does not respond at all. I have already updated and reinstalled the app and restarted my phone, but nothing has helped. Could you please check and help me enable Local mode?
Title: Korean voice input does not work in Omi Chat, while Korean live transcription works normally Description: I am using an Omi CV1 with the Android app on a Galaxy S25 Edge. Korean transcription works normally during continuous/live conversation recording. The device detects my Korean speech and creates Korean transcripts and Conversations correctly. However, Korean voice input does not work inside Omi Chat. When I tap the microphone in Omi Chat and speak Korean, the app detects that I spoke, but then shows: “Speech was detected, but the transcription provider returned no transcript.” This happens repeatedly with Korean. If I use English voice input in the exact same Omi Chat interface, transcription works and Omi replies normally. So the issue appears to be specific to the voice transcription path used by Omi Chat for Korean, not the microphone, Bluetooth connection, or the device’s general Korean transcription support. Steps to reproduce: Connect an Omi CV1 to the Android Omi app. Confirm that normal/live conversation recording successfully transcribes Korean. Open Omi Chat. Tap the microphone button. Speak a clear Korean sentence. Omi detects speech but returns: “Speech was detected, but the transcription provider returned no transcript.” Repeat the same test in English. English voice transcription works and the assistant responds normally. Expected behavior: Korean voice questions in Omi Chat should be transcribed and sent to the assistant, just like English voice questions.
Please add Android immersive mode or an in-app option to hide Samsung’s 3-button navigation bar. On my Samsung phone, the navigation buttons stay visible and overlap Omi’s bottom controls. Ideally, Omi should either hide the bar or automatically add bottom padding.
My new Omi turns on when I push it for 3 seconds, it vibrates and connects to the app. Then shortly after it vibrates and turns off. Sometimes when it connects it says 1% battery and seconds later it says 100% battery.
So far my experience has not been good. The first Omi I ordered did not work and I had to wait two weeks for a replacement. Now I have the replacement and the mind app does not work.
Exception: Failed to rebuild knowledge graph: {“detail”:”Canonical knowledge graph state is derived from canonical memories and cannot be deleted or rebuilt directly.”} It worked for the first weeks, than stopped out of nowhere. I have the “pro” license.
No Voice + No Transcription after firmware 3.0.21 on Omi CV1