Emotions.Tech can add emotion to text to speech.
Imagine Siri reading you a poem in her stilted mostly monotone voice.
Now imagine that she knew when to sound happy, when to sound sad, when to giggle and when to be surprised. This can be done easily and effectively, because Emotions.tech can provide word-by-word analysis like this:
Using this information Siri's voice could subtly modulate while she's speaking. Practically, this can be implemented re-recording the original voice spoken with different tones.
To realise this idea, we need a partner in the TTS field. Please contact us if you would like to work with us on this.