6th of 26 voices at starting to talk fast
Non-streaming latency (median): 194 ms
Deepgram Flux: 6th of 26 voices at starting to talk fast. Its weakest is naturalness, 26th of 26.
Last updated Methodology by Dileep Chagam
Where it ranks among all 26 voices, in plain words.
The first read of each test sentence, unedited.
Pick any voice to set against it. Every pair has its own page.
Deepgram Flux starts in 194 ms on a typical sentence (median) and 242 ms on the slowest 1 in 10 (p90), 6th of 26 voices.
Deepgram Flux starts in 194 ms on a typical sentence (median) and 242 ms on the slowest 1 in 10 (p90), 6th of 26 voices.
Yes. With streamed text, Deepgram Flux starts 154 ms after the last word arrives.
Yes. After the stop it still sends 80 ms of audio (median), which the agent throws away.
Deepgram Flux read 77.4% of the hard parts right, as speech-to-text heard it. Its weakest kind is read from context, at 68.8%. This is an estimate, not a listening test.
Deepgram Flux lists at $45.00 per million characters.