gangster_dave

I completely understand. This product is in beta testing with no paid customers. We'll definitely have that in there at launch.

That's exactly right. The aim of this product is to enable businesses to offer that service to their users. Right now, you can only get intelligent dictation by having users already come to the table with some of these apps, like Wispr Flow.

I don't think you can do that quite yet, since the TTS APIs require a full phrase in order to output fluent sounding speech. If the input is short, then the delivery/emotion/pauses are random per word/token. I actually think that type of system will be possible once we have a multimodal model that understands and outputs speech, with the intelligence of GPT4.

The latency is dependent on those three APIs, but the biggest bottleneck is the GPT4 API. Its latency varies throughout the day, from <200ms to >1s. There are several application-level optimizations in Vibrato, like managing streaming audio and streaming text, but these aren't as impactful as the API latencies.

Great feedback—will add tomorrow!

Thanks for the feedback! The product is still very, very early, so we don't have a lot of marketing or documentation. We'd be pushing towards a Jupyter notebook interface for non-coding tasks, though. We've discussed adding a code interface (through blocks), so maybe that's something to target next. Again, thanks!

Could you do boosting with Typesense, like favoring more recent results?