Sadly I don't think any (commercial / phoneme based) TTS would be very listenable for a podcast. Those are hand rolled fragments of speech. ( Think old school Satnavs "In " + "30 yards " + "turn left"
Will clean it up if there's enough interest.
It's in a few parts ... 'transcript' generation, then the speech assembly... then dropping in the backing track + intro / outro, then uploading.
Sadly I don't think any (commercial / phoneme based) TTS would be very listenable for a podcast. Those are hand rolled fragments of speech. ( Think old school Satnavs "In " + "30 yards " + "turn left"