Left is faster. Higher reads more hard parts right.
Deepgram Flux is better on stopping. Inworld TTS-2 Flash is better on non-streaming latency, streaming latency, accuracy, naturalness and price.
Last updated Methodology by Dileep Chagam
Measure by measure, the same sentences, the same place.
An estimate: speech-to-text transcribers scored it, not people.
The same sentence, read by each voice.
Inworld TTS-2 Flash. Deepgram Flux: 194 ms. Inworld TTS-2 Flash: 46 ms.
Inworld TTS-2 Flash. Deepgram Flux: 194 ms. Inworld TTS-2 Flash: 46 ms.
Inworld TTS-2 Flash. Deepgram Flux: 154 ms. Inworld TTS-2 Flash: 45 ms.
Deepgram Flux. Deepgram Flux: 80 ms. Inworld TTS-2 Flash: No stop message.
Inworld TTS-2 Flash. Deepgram Flux: 77.4%. Inworld TTS-2 Flash: 93.2%.
Inworld TTS-2 Flash. Deepgram Flux: $45.00. Inworld TTS-2 Flash: $15.00.