The Marchex 2018 English Conversational Telephone Speech Recognition System
This is an incremental improvement for Marchex's call analytics system, enhancing performance in natural language processing pipelines.
The paper tackles improving speech recognition for spontaneous customer-to-business telephone conversations by employing a semi-supervised lattice-free maximum mutual information training process, resulting in a 3.3% absolute reduction in word error rate and 3x faster decoding speed over the previous system.
In this paper, we describe recent performance improvements to the production Marchex speech recognition system for our spontaneous customer-to-business telephone conversations. In our previous work, we focused on in-domain language and acoustic model training. In this work we employ state-of-the-art semi-supervised lattice-free maximum mutual information (LF-MMI) training process which can supervise over full lattices from unlabeled audio. On Marchex English (ME), a modern evaluation set of conversational North American English, we observed a 3.3% (3.2% for agent, 3.6% for caller) reduction in absolute word error rate (WER) with 3x faster decoding speed over the performance of the 2017 production system. We expect this improvement boost Marchex Call Analytics system performance especially for natural language processing pipeline.