Feature Ideas

Textual transcripts lack the emotional delta present in vocal delivery. I want audio-native interpretation to capture sarcasm, excitement, and friction that words alone mask. Specifications: Tonal Distribution: Provide a percentage-based breakdown of emotional states (e.g., 40% Engagement, 20% Sarcasm, 10% Reticence).Temporal Markers: Identify specific vectors where the emotional tone shifts, noted by precise timestamps (e.g., "Shift to annoyance at 08:42").Acoustic Inference: Analyze pitch, cadence, and inflection to synthesize intent rather than just transcribing syntax. Objective: Transform transcripts from flat records into high-fidelity intelligence.

It would be really awesome if we could configure an access point for pocket to connect to (e.g., your home’s wireless router) for auto syncing to pocket cloud instead of directly to your phone via the app. This would be significantly easier, faster, and additionally an option to have it always sync or on while charging would be incredible! Yeah, battery life would probably take a hit but worth it in my opinion. I saw other idea posts here but they seemed to just be talking about connecting the device to the mobile app via WiFi and NOT the pocket device itself connecting to an access point. Unless I’m missing some aspect of security or hardware limitations, this seems like it would benefit a lot of pocket users who aren’t around their phone all day.

I’m currently testing Pocket intensively and already see a lot of value for my daily work, especially as a central place for thoughts, notes and action items. One feature would make Pocket significantly easier for me to integrate into my daily workflow: support for AirPods or other Bluetooth headsets as the audio input. I travel a lot and frequently work from my car, airports and airplanes, where I almost exclusively use AirPods. Currently, I need to hold or position the Pocket device itself to record something. This is inconvenient while travelling and, particularly while driving, not ideal from a safety perspective. My ideal workflow would be very simple: press the button on Pocket to start a recording, while the actual audio is captured through the microphone of my connected AirPods. Being able to record phone calls conducted through AirPods would also be extremely valuable, although I assume call recording is already a known use case. Is support for Bluetooth microphones / AirPods something that is currently on your roadmap?

Seems like mindmaps are more eye candy than helpful on most notes. I'd like to be able to opt-in to adding one over having it default on for every note.

Description: I am requesting a dedicated input field that appears immediately after a recording is completed, but before the transcription and summarization process begins. This field should allow users to provide: Names of participants. The specific context of the conversation. Relevant technical terminology or additional background information. Use Case: By providing this context upfront, the AI can produce a significantly more accurate transcript and summary. This eliminates the need for manual editing after the fact and ensures that the output is immediately ready for professional workflows.

Right now, when you sync, there's a button you click to generate a summary and a pop-up happens asking you to choose a template. I feel like here it should also have you label the speakers, so that when it's summarizing it will summarize with names instead of "speaker 3". In order to change it from "speaker 3" to "john" you have to go into the transcript and tap on them. It would make a lot more sense to be able to access the speaker labeling functions quicker and easier.

It’s possible to bump the record button or even click to record and then change your mind and cancel before any content is recorded. The system still processes and saves this. It would be good to have a setting that automatically deletes this if no content was recorded or a recording is less than “0” seconds long. For future hardware, it might be good to have a toggle lock to avoided accidental button presses.

How about a companion Watch app, with a watch complication that would allow you to start recording a meeting by tapping the complication on the watch, rather than taking out your phone and using the app. In some meetings with execs or clients, they don't want to see you using your smartphone. Using a watch would provide a more discreet way to start your recording.

Timeout for the LED status light after say 10 minutes of no use. It can draw unnecessary attention and can be a disturbance at night even though it isn't super bright.

It would save me many hours if I could swipe a conversation left to on the dashboard to delete accidental recordings, of which I have at least a handful every day.

When adding recording from a file to the app, detect the recording date and time automatically from the metadata.

So you don't need to open the app, but it syncs in the background. Would lower friction so much. Loved the idea about not needing to interact with an app, but isn't there a 100%. My oura and my whoop can do background syncing over Bluetooth, should be possible for the pocket also. You just leave it on for some time after the recording, so that it can sync. A notification when it's done, so you know when you can turn it off.

I need the ability to rename or correct the name of a recording before it uploads the file into OneDrive.

I love the add attachment feature, use it often. But it’s only available during the meeting and after summarizing. I want to add attachment also after transferring from Pocket to the app, so the summary can take into consideration the attachments. That would stop the need to redo the summary in many cases.

Most of my meetings are virtual and I'm wearing a headset. This makes pocket way less useful for me as it's unable to record meetings that are not happening live or if it cannot hear the other people speak. A pocket app that can listen in on meetings or be able to import transcripts would be awesome