I bought the top tier and have been messing around with Speechki, but today I did my first serious task with it, producing video narration for my company's marketing team. I was BLOWN AWAY by the nuance in the voice! I really didn't think the technology was capable of this, but my ear could really not discern that this was anything other than a professional human voice actor who understood what she was talking about.

I'm not offended that all the features aren't there yet, like editing, correcting phonemes, adding emotion, creating dialogue, etc. I trust that they're coming, since the developers have said as much. But even generating audio using the current basic slot-machine-type process, I'm getting results that obliterate any other product I've tried (and maxed out on here on AppSumo -- ouch!).

The one extra feature that is here, which should be dropped immediately, is the music generator. At present, it's comically awful: It produces up to a minute and a half of music, but when I asked it to create "ambient, quiet music for studying or sleeping", it came up with something kind of New Agey that gained drumbeats as it gained momentum, then escalated into a sound that I can only describe as the screaming whistle of kettle. (This happened with two separate attempts to generate music.) It's horrifying, unusable, and feels like a distraction from the main product.

In my mind, there's this and ElevenLabs, but the latter is pay-as-you-go, whereas it's pretty exciting to imagine the possibilities of Speechki's unlimited level.

Thanks to the developers for creating this great tool. I look forward to seeing how it grows.