5 ms·
I've had difficulty obtaining useful results from the smaller (7B-sized) models. The issue lies in the content, not the speed. If you could stream the text-to-s
by schappim 4y ago
I've had difficulty obtaining useful results from the smaller (7B-sized) models. The issue lies in the content, not the speed. If you could stream the text-to-speech, the speed alone would be satisfactory.
- amusedcyclist 4y agoYou're right I overestimated how fast we talk!
- fwlr 4y agoSome rules of thumb I use for estimating this kind of stuff 100wpm: Max typing speed 200wpm: Max speaking speed 300wpm: Max listening speed, max reading speed with subvocalisation 900wpm: Max reading speed without subvocalisation
- mromanuk 4y agoDoin napkin math, this model should be hitting 900wpm