Left is faster. Higher reads more hard parts right.
Gemini 3.8 Flash-Lite is better on non-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.
Gemini 3.8 Flash-Lite. Gemini 3.8 Flash-Lite: 729 ms. MAI-Voice-2.1: 861 ms.
Gemini 3.8 Flash-Lite. Gemini 3.8 Flash-Lite: 729 ms. MAI-Voice-2.1: 861 ms.
Neither has a figure. Gemini 3.8 Flash-Lite: Can't stream. MAI-Voice-2.1: Can't stream.
Neither has a figure. Gemini 3.8 Flash-Lite: No stop message. MAI-Voice-2.1: No stop message.
Gemini 3.8 Flash-Lite. Gemini 3.8 Flash-Lite: 88.6%. MAI-Voice-2.1: 82.2%.
Gemini 3.8 Flash-Lite. Gemini 3.8 Flash-Lite: $19.03. MAI-Voice-2.1: $22.00.