Computing Optimal Location of Microphone for Improved Speech Recognition
This work addresses microphone positioning for improved speech recognition accuracy, but it is incremental as it extends prior research on measurement errors.
The paper tackles the problem of microphone placement for speech recognition by identifying the optimal location given an erroneous initial position, using Monte-Carlo simulations to select positions that maximize performance on a general-purpose ASR system, showing that the optimal location is unique and noise-dependent.
It was shown in our earlier work that the measurement error in the microphone position affected the room impulse response (RIR) which in turn affected the single-channel close microphone and multi-channel distant microphone speech recognition. In this paper, as an extension, we systematically study to identify the optimal location of the microphone, given an approximate and hence erroneous location of the microphone in 3D space. The primary idea is to use Monte-Carlo technique to generate a large number of random microphone positions around the erroneous microphone position and select the microphone position that results in the best performance of a general purpose automatic speech recognition (gp-asr). We experiment with clean and noisy speech and show that the optimal location of the microphone is unique and is affected by noise.