mlsanityData to 14 Oct 2026

DCTX-Conformer: Dynamic context carry-over for low latency unified streaming and non-streaming Conformer ASR

arXiv 2306.08175 ↗First posted 13 Jun 2023eess.AS cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Conformer-based end-to-end models have become ubiquitous these days and are commonly used in both streaming and non-streaming automatic speech recognition (ASR). Techniques like dual-mode and dynamic chunk training helped unify streaming and non-streaming systems. However, there remains a performance gap between streaming with a full and limited past context. To address this issue, we propose the integration of a novel dynamic contextual carry-over mechanism in a state-of-the-art (SOTA) unified ASR system. Our proposed dynamic context Conformer (DCTX-Conformer) utilizes a non-overlapping contextual carry-over mechanism that takes into account both the left context of a chunk and one or more preceding context embeddings. We outperform the SOTA by a relative 25.0% word error rate, with a negligible latency impact due to the additional context embeddings.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 27 of its 37 references that are in the map. First posted on arXiv 13 Jun 2023.

Builds on

27 of 37 references.

References · 27

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
  4. 2013Speech Recognition with Deep Recurrent Neural Networkscited by 669
  5. 2018ESPnet: End-to-End Speech Processing Toolkitcited by 558
  6. 2015Listen, Attend and Spellcited by 543
  7. 2021VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretationcited by 463
  8. 2015Deep Speech 2: End-to-End Speech Recognition in English and Mandarincited by 451
  9. 2015Attention-Based Models for Speech Recognitioncited by 401
  10. 2016Joint CTC-Attention based End-to-End Speech Recognition using Multi-task Learningcited by 257
  11. 2020Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Losscited by 235
  12. 2021WeNet: Production oriented Streaming and Non-streaming End-to-End Speech Recognition Toolkitcited by 125
15 more
  1. 2020Emformer: Efficient Memory Transformer Based Acoustic Model For Low Latency Streaming Speech Recognitioncited by 113
  2. 2017Monotonic Chunkwise Attentioncited by 83
  3. 2019Two-Pass End-to-End Speech Recognitioncited by 64
  4. 2020Deliberation Model Based Two-Pass End-to-End Speech Recognitioncited by 57
  5. 2020Unified Streaming and Non-streaming Two-pass End-to-end Model for Speech Recognitioncited by 39
  6. 2020Streaming Transformer-based Acoustic Models Using Self-attention with Augmented Memorycited by 35
  7. 2019Self-Attention Aligner: A Latency-Control End-to-End Model for ASR Using Self-Attention Network and Chunk-Hoppingcited by 33
  8. 2019Transformer ASR with Contextual Block Processingcited by 28
  9. 2020Transformer-based Online CTC/attention End-to-End Speech Recognition Architecturecited by 28
  10. 2020Transformer Transducer: One Model Unifying Streaming and Non-streaming Speech Recognitioncited by 24
  11. 2020Conv-Transformer Transducer: Low Latency, Low Frame Rate, Streamable End-to-End Speech Recognitioncited by 23
  12. 2021Dual Causal/Non-Causal Self-Attention for Streaming End-to-End Speech Recognitioncited by 14
  13. 2022CUSIDE: Chunking, Simulating Future Context and Decoding for Streaming ASRcited by 11
  14. 2020Weak-Attention Suppression For Transformer Based Speech Recognitioncited by 7
  15. 2023Dynamic Chunk Convolution for Unified Streaming and Non-Streaming Conformer ASRcited by 7

Built on by

Not cited yet.