The benchmark is using the latest stable of DS >>> https://github.com/Picovoice/speech-to-text-benchmark/blob/master/requirements.txt#L3 https://github.com/Pico

The benchmark is using the latest stable of DS >>> https://github.com/Picovoice/speech-to-text-benchmark/blob/master/requirements.txt#L3 https://github.com/Picovoice/speech-to-text-benchmark/blob/m... the data for cloud-based is from 2022.