Minor bugs around summaries

[sidd_heypocket]

I dont know if this is the right place to post. But i am observing two things:

1. Even after i update transcript to assign names to speakers, the summaries sometimes have Speaker 0, Speaker 1, etc, thats annoying. And in some cases, even after I have tagged speakers, it shows me as following (which makes no sense)

Sidd (Dean) – Client feedback aggregation
Dean (Ryan) – Operational and GTM alignment

2. I set my default mode in settings to be AutoPilot and the model to be GPT-5. But it still generates summaries on Auto-detect, i have to go toggle to Auto-Pilot and re-generate the summary.

[mhirrrr]

hey @sidd_heypocket you can follow this

Speaker identification: Transcript is correct, Summary is not

hey guys! given current transcript generation limitations we are trying to solve this issue to the best of our ability so if you guys do not mind sharing the details of the recording where you faced this issue to hey@heypocket.com. If would give us more data points to work with so we can make this the best experience possible for everyone. thanks, mihir

we are currently working on improvements to make sure this doesn’t happen (making the process more predictable)

attaching @chay to look at the second issue for you

[sidd_heypocket]

This is still an issue? What details do you need of the recording?

[mhirrrr]

hey sidd as mentioned on the other post we are trying to work with the models to get more accurate summaries around speakers and research is being done around that so please allow us some time

[mhirrrr]

thanks to @sidd_heypocket we were able to make some steps towards improving speaker identification! if you regenerate your summaries you should find the tagging a lot better thanks!