Spellar AI Reviews

Was a bit apprehensive about giving this a punt after reading the negative reviews, but really glad I did. As I don't use an iPhone, the lack of call recording on that device doesn't impact me.

Once I'd set up my own api key, it has been super useful, discreet and works seamlessly with my workflow. I can integrate it into Notion and from there into Claude to automate, extracting info, creating to-do lists, etc so it ticks most of the boxes that I need it for.

It's very simple, and I prefer it to Fathom, which had been my go-to, but the meeting bot was becoming a distraction.

It's simple, has been reliable for me, and I hope they continue to develop the product and perhaps even create an Android app.

I only use the app on desktop (Mac) for online meetings and like the fact that it's not tied to an app (Zoom, Meet, Slack, etc.). The downside here is that it doesn't know who the people in the meeting are (maybe there's some integration, but haven't set that up).

My main thing is that the meetings I had were not in English (they were in Portuguese) and the app states that the recording is noisy. When I play the recording, it's very clear.

I didn't see:
* How to tell the app the language the meeting is in.
* How to re-analyze the audio with more input: language, speaker names, etc.
* How to add new Actions after the meeting.

Liked:
* Easy to start recording.
* Mostly simple UI, although a bit cluttered in adding the meeting to Notion, etc. that could be a drop down.
* Saves the meeting recording, in case we need to play back.

Note: Haven't tried any integrations (calendar, tasks, notes, etc.)

At first glance, the product appears promising and positioned as a tool to streamline and extend meeting minutes. However, in its current state, it does not deliver on its core promise.

The overall experience feels more like an early alpha release than a stable, production-ready solution. Core functionality—specifically generating usable meeting minutes after a session—does not work reliably. In practice, the product consistently fails with errors, and the support and bug reporting sections are filled with similar issue reports from other users.

Although a recent patch has been released, the underlying problems persist, at least in my case. I reached out to support to clarify the issues, but did not receive any response.

From a business perspective, this raises concerns not only about product maturity but also about vendor reliability. A tool that fails at its primary use case and lacks responsive support cannot be considered viable for professional use.

As long as these issues remain unresolved, I would not recommend purchasing or relying on this product. There are more stable alternatives available that fulfill the same purpose.

Unfortnately, I had the same experience as many others here. I never figured out how to set it up to give me a summary. The first 3 reviews with 5 stars seem to be incentivized, as all reviews after that are 3 stars or lower. To be honest: the product sounds quite promising, but the fact, that so many users did not reach to get it working (including me) is just worth 1 star (or even 0).

I was never really able to get this to work. I had a real problem recording audio and was never able to get it to summarize anything, even if I could get a transcript at all. It did a poor job of detecting and identifying speakers other than myself.