Streaming Inverse Text Normalization (ITN) is vital for converting spoken-form outputs from streaming Automatic Speech Recognition into formatted written text.
Existing streaming ITN methods rely on hybrid systems combining neural tagging with expert-crafted finite-state transducer rules, limiting scalability across domains and languages.
While end-to-end models offer superior scalability by learning directly from data, standard encoder-decoder architectures are inherently non-streaming due to global attention.
In This Paper
In this paper, we propose an efficient streaming end-to-end ITN system, adapting a pretrained text-to-text model to leverage its robust linguistic knowledge.
To enable streaming, we introduce architectural adaptations, a specialized training strategy, a Read-Tag-Write decoding policy, and inference optimizations.
Experiments
Experiments on Vietnamese datasets show accuracy comparable to non-streaming baselines, outperforming hybrid methods while satisfying real-time latency requirements.