Researchers at Carnegie Mellon University are using custom computer chips to tackle a problem in speech recognition that software largely hasn’t been able to solve.