Speech Recognition system in C and C++, built using windows core audio APIs for audio capturing from audio endpoints and , Whisper.cpp for processing the audio and transcribing the text as output.
Although this is preliminary work, the program currently processes audio for a duration of 10 seconds, which can be adjusted as needed. However, the streaming of real-time audio is still pending implementation, which will be done in few days.
The models for processing audio can be downloaded from https://huggingface.co/ggerganov/whisper.cpp/tree/main