For a product my company is developing, we are using the ADAU1761 codec and two Wolfson MEMS microphones to eventually do speech recognition. Unfortunately, the audio quality received is not as high as we'd hoped due to the casing.
At the moment we are not utilizing the DSP's features, but piping the analog input directly to the Speech-to-Text engines. This works pretty good <2m distance, but from there the recognition quality drops fast.
At my company we have not much experience with DSP configuration, and frankly, it's not a job that seems to be easily findable. So I'm hoping to find someone who can freelance to help us out.
How would a person that can optimize the DSP's configuration call himself, and where do they usually work?
Thanks a lot in advance,