Towards Improved Room Impulse Response Estimation for Speech Recognition
This work addresses the challenge of accurate RIR estimation for enhancing speech recognition in reverberant environments, representing an incremental improvement over existing methods.
The paper tackles the problem of blind room impulse response (RIR) estimation to improve far-field automatic speech recognition (ASR), showing that their GAN-based model outperforms state-of-the-art baselines by 17% on an energy decay relief metric, 22% on an early-reflection energy metric, and reduces word error rate by 6.9% in ASR evaluation.
We propose a novel approach for blind room impulse response (RIR) estimation systems in the context of a downstream application scenario, far-field automatic speech recognition (ASR). We first draw the connection between improved RIR estimation and improved ASR performance, as a means of evaluating neural RIR estimators. We then propose a generative adversarial network (GAN) based architecture that encodes RIR features from reverberant speech and constructs an RIR from the encoded features, and uses a novel energy decay relief loss to optimize for capturing energy-based properties of the input reverberant speech. We show that our model outperforms the state-of-the-art baselines on acoustic benchmarks (by 17\% on the energy decay relief and 22\% on an early-reflection energy metric), as well as in an ASR evaluation task (by 6.9\% in word error rate).