FAST-RIR: Fast neural diffuse room impulse response generator
This work addresses the need for efficient RIR generation in speech processing applications, offering a fast alternative to existing methods, though it is incremental in improving speed and accuracy.
The paper tackles the problem of generating room impulse responses (RIRs) for acoustic environments by introducing FAST-RIR, a neural-network-based generator that achieves an average error of 0.02s for reverberation time and is 400 times faster than a state-of-the-art simulator on CPU while maintaining similar performance in automatic speech recognition (ASR) tasks.
We present a neural-network-based fast diffuse room impulse response generator (FAST-RIR) for generating room impulse responses (RIRs) for a given acoustic environment. Our FAST-RIR takes rectangular room dimensions, listener and speaker positions, and reverberation time as inputs and generates specular and diffuse reflections for a given acoustic environment. Our FAST-RIR is capable of generating RIRs for a given input reverberation time with an average error of 0.02s. We evaluate our generated RIRs in automatic speech recognition (ASR) applications using Google Speech API, Microsoft Speech API, and Kaldi tools. We show that our proposed FAST-RIR with batch size 1 is 400 times faster than a state-of-the-art diffuse acoustic simulator (DAS) on a CPU and gives similar performance to DAS in ASR experiments. Our FAST-RIR is 12 times faster than an existing GPU-based RIR generator (gpuRIR). We show that our FAST-RIR outperforms gpuRIR by 2.5% in an AMI far-field ASR benchmark.