Actually, my question should be how do I name the speakers because that’s not happening right now. So I guess it’s a two-part question:
I’m on the three month pro trial and I’m wondering why it’s not doing that, can someone update me please?
And since it’s not doing that, I would like to name the speakers manually, can you please tell me how to do that in the app because I don’t see an option.
Hey @vbrown2025 could you just mail us at hey@heypocket.con with an example of the recording where you are facing this so we can investigate
so currently we do our best to do speaker matching and recognition and are the best in the market we are also making a bunch of improvements ro make this experience better.
Yeah that’s fine. I’ve got about 4 new ones today to upload after this meeting. So in about 20 mins you’ll see a fresh batch. But go ahead and check it out after that.
I’m not sure, but it seems that some of the Summary Themes in Settings don’t label the speakers (Speaker 1, Speaker 2, etc.), but when I changed it back to the Auto-Detect, it at least captured “Speaker 1, Speaker 2, etc.) and then it was fairly easy to identify the speakers and have their names show up for when they were speaking. That’s just my experience so far, so I could be wrong. I don’t know if that’s how it’s supposed to work on not.
I will do that, but as Nir said, I’ve never seen any speaker identification at all. I was also assuming it was coming out in the next update. I will email you.
Hey @vbrown2025 the algorithm works if you have previously tagged users you can mail us if there are cases this hasn’t happened and we can provide you more information
I’m having similar issues. It won’t allow me to rename speakers throughout the transcript. Once I add a new speaker it automatically changes every speaker in the transcript to that name. And any speaker name I previously added is removed from the list of speakers and only the latest one I added is listed to select from.
Is this a known issue that is working to be resolved?
@mhirrrr I am also having problems with basic speaker labeling. In the transcripts, many male voices have been identified as me and auto labeled as such. When I click on the conversation and attempt to add a new speaker label, I type in the new speaker name, click the + button, or click the “return” button, the new label will NOT apply and update accordingly. All of the labels continue to be stuck under my name. This is going to be a HUGE problem if I can’t remedy this fast, as I already have hours of transcripts from this week alone to update tags from. Help please!
@mhirrrr@pete-w@Akshay I want to say that we really appreciate the hard work you’re putting in behind the scenes. I understand that the issue with speaker identification – particularly regarding how changes affect speaker names globally and the inability to rename speakers – has been raised multiple times. I completely trust that it’s on your radar, and I hope it can be prioritized for resolution.
The global “Ask Pocket” feature is a fantastic addition that we’re eagerly looking forward to. Heck, it appears that’s one of the main reasons why most of us got the device (and why it’s the most voted for feature). Still, I believe that it’s essential to address these concerns about renaming speakers so they don’t get overlooked.
I’m hoping we can keep the positive momentum going. I can sense some users’ frustration on this matter, and I wanted to bring it to your attention to ensure it gets the focus it deserves. Thank you for your continued efforts.
hey @p_odoi thanks for this. rest assured we have a lot in store for users the next few weeks.
good news is we have figured out a fix for this tagging issue and are working with Apple and Google to get this live so we can see if this fixes it for everyone