Left is faster. Higher reads more hard parts right.
Deepgram Aura-2 is better on streaming latency and naturalness. Soniox TTS RT v2 is better on non-streaming latency, stopping, accuracy 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.
Soniox TTS RT v2. Deepgram Aura-2: 281 ms. Soniox TTS RT v2: 241 ms.
Soniox TTS RT v2. Deepgram Aura-2: 281 ms. Soniox TTS RT v2: 241 ms.
Deepgram Aura-2. Deepgram Aura-2: 126 ms. Soniox TTS RT v2: 183 ms.
Soniox TTS RT v2. Deepgram Aura-2: 40 ms. Soniox TTS RT v2: 0 ms.
Soniox TTS RT v2. Deepgram Aura-2: 77.2%. Soniox TTS RT v2: 83.6%.
Soniox TTS RT v2. Deepgram Aura-2: $30.00. Soniox TTS RT v2: $14.81.