Hybrid convo voice applications?
Hi Leor,
Absolutely a good point!
A technology that reads the text for tone and can then automatically generate SSML code for the TTS.
yes, the reading software could understand context and “sentiment”. it doesn’t seems a so difficult goal and I foresee next generation TTS APIs will cover the “expression” lack in today tech.
Waiting for this evolution (coming very soon immo), in the meantime a possible way is having what I call “hybrid” (made by humans and by bots) convo apps:
Imagine a sort of interactive audiobook, where content is prerecorded (let say a podcast), made by humans (voice actors).
Spoken editions + Soundscapes + metabot voiceover
On top of this podcast, a metabot act as voiceover/listener assistant (completing linear prerecorded narrative with extra metadata info).
What do you think about it?
I’m experimenting such realms…