CAT: CRF-based ASR Toolkit
This toolkit addresses the need for efficient and flexible end-to-end speech recognition tools for researchers and practitioners, though it is incremental as it builds on existing CRF and CTC methods.
The authors introduced CAT, an open-source toolkit for automatic speech recognition that uses a CRF-based framework with CTC-inspired state topology, achieving state-of-the-art results on benchmarks like Switchboard and Aishell with fewer parameters and competitive performance against hybrid models.
In this paper, we present a new open source toolkit for automatic speech recognition (ASR), named CAT (CRF-based ASR Toolkit). A key feature of CAT is discriminative training in the framework of conditional random field (CRF), particularly with connectionist temporal classification (CTC) inspired state topology. CAT contains a full-fledged implementation of CTC-CRF and provides a complete workflow for CRF-based end-to-end speech recognition. Evaluation results on Chinese and English benchmarks such as Switchboard and Aishell show that CAT obtains the state-of-the-art results among existing end-to-end models with less parameters, and is competitive compared with the hybrid DNN-HMM models. Towards flexibility, we show that i-vector based speaker-adapted recognition and latency control mechanism can be explored easily and effectively in CAT. We hope CAT, especially the CRF-based framework and software, will be of broad interest to the community, and can be further explored and improved.