Computing Optimal Location of Microphone for Improved Speech Recognition

TL;DR

This paper investigates how to determine the best microphone placement for speech recognition by using Monte-Carlo simulations to account for positional errors, improving recognition accuracy in various noise conditions.

Contribution

It introduces a systematic method to find the optimal microphone location considering positional errors, enhancing speech recognition performance.

Findings

Optimal microphone location is unique for given conditions.

Noise influences the optimal placement.

Monte-Carlo technique effectively identifies the best position.

Abstract

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.