Dessa: Conversation Model trained only on speech

I'm surprised it took this long! Clearly the next step is joint training on text and audio. The speech realism is very good, but the intelligence is pretty clearly 2024-level.

Those demos are, omg, like, the worst, donchya think?

crazy, if true!

[flagged]

[dead]

[flagged]

[dead]

[flagged]

Dessa is written in a swirly feminine font.

There's no male or neutral voice.

All examples are of white male software engineers talking to what I assume will be ultimately marketed as a 'girlfriend' bot (or used for love scams).

The single paragraph hand waving at the importance of 'safe AI empathy' doesn't dissuade me at all.

Sell your idea as what it is, yet another 'perfect' woman shaped object a hypothetical man can own and use how he likes.

Sorry I'm lost. The sample on the homepage sounds absolutely awful. Not one bit of the cadence individually and together, the tonal ranges in responses or the overall structure and feel is comfortable to listen to, it's so far off passing as 'real' that I feel like I must be missing something.

Authentic to what? That annoying clinger who loves interrupting you when you're trying to speak in turns?

Is having a duplex model constantly cutting in with uninteresting quips something people are looking for?