Multilingual speech-to-text benchmark for transcribe.cpp models on a laptop
Site comparing accuracy and speed of speech-to-text models supported by transcribe.cpp, measured per language on FLEURS, with data from tests on a 4750U laptop.
A website presents a multilingual benchmark of the accuracy and speed of speech-to-text models supported by transcribe.cpp. Measurements are reported per language, using the FLEURS dataset. According to the source, the performance data come from tests on a laptop CPU/GPU identified as 4750U.
The description states that the tool helps compare local ASR models on accuracy and speed across several languages before choosing one. The source does not give further details such as measurement dates or specific results. To verify these points, consult the original site and the methodology it publishes.