Left is faster. Higher reads more hard parts right.
Fish Audio S2.1-Pro is better on non-streaming latency, streaming latency and accuracy. Boson Higgs TTS 3 is better on naturalness.
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.
Fish Audio S2.1-Pro. Fish Audio S2.1-Pro: 333 ms. Boson Higgs TTS 3: 472 ms.
Fish Audio S2.1-Pro. Fish Audio S2.1-Pro: 333 ms. Boson Higgs TTS 3: 472 ms.
Fish Audio S2.1-Pro. Fish Audio S2.1-Pro: 321 ms. Boson Higgs TTS 3: Can't stream.
Neither has a figure. Fish Audio S2.1-Pro: No stop message. Boson Higgs TTS 3: No stop message.
Fish Audio S2.1-Pro. Fish Audio S2.1-Pro: 92.0%. Boson Higgs TTS 3: 88.8%.
They are level. Fish Audio S2.1-Pro: $15.00. Boson Higgs TTS 3: $15.00.
Cekura runs these simulated calls on your agent and scores every one.