# Semantic Segmentation with Bidirectional Language Models Improves Long-form ASR

arXiv: https://arxiv.org/abs/2305.18419 · HTML: https://mlsanity.com/paper/2305.18419/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Speech and audio distillation, one of 12 directions in Knowledge distillation.
- Builds on 15 of its 23 references that are in the map.
- First posted on arXiv 28 May 2023.

## Placed in

- Knowledge distillation / [Speech and audio distillation](https://mlsanity.com/distill/speech-and-audio-distillation/)

## Builds on (most cited first, 15 of 15)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [Lingvo: a Modular and Scalable Framework for Sequence-to-Sequence Modeling](https://mlsanity.com/paper/1902.08295/) (2019)
- [FastEmit: Low-latency Streaming ASR with Sequence-level Emission Regularization](https://mlsanity.com/paper/2010.11148/) (2020)
- [Less Is More: Improved RNN-T Decoding Using Limited Label Context and Path Merging](https://mlsanity.com/paper/2012.06749/) (2020)
- [Dissecting User-Perceived Latency of On-Device E2E Speech Recognition](https://mlsanity.com/paper/2104.02207/) (2021)
- [E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR](https://mlsanity.com/paper/2204.10749/) (2022)
- [Turn-Taking Prediction for Natural Conversational Speech](https://mlsanity.com/paper/2208.13321/) (2022)
- [A Unified Cascaded Encoder ASR Model for Dynamic Model Sizes](https://mlsanity.com/paper/2204.06164/) (2022)
- [Directed Speech Separation for Automatic Speech Recognition of Long Form Conversational Speech](https://mlsanity.com/paper/2112.05863/) (2021)
- [Smart Speech Segmentation using Acousto-Linguistic Features with look-ahead](https://mlsanity.com/paper/2210.14446/) (2022)
- [E2E Segmentation in a Two-Pass Cascaded Encoder ASR Model](https://mlsanity.com/paper/2211.15432/) (2022)
- [FullStop:Punctuation and Segmentation Prediction for Dutch with Transformers](https://mlsanity.com/paper/2301.03319/) (2023)
- [Streaming Punctuation: A Novel Punctuation Technique Leveraging Bidirectional Context for Continuous Speech Recognition](https://mlsanity.com/paper/2301.03819/) (2023)
- [Improved Long-Form Spoken Language Translation with Large Language Models](https://mlsanity.com/paper/2212.09895/) (2022)

## Built on by (most cited first, 2 of 2)

- [Multilingual and Fully Non-Autoregressive ASR with Large Language Model Fusion: A Comprehensive Study](https://mlsanity.com/paper/2401.12789/) (2024)
- [DOA: Training-Free Decoder-Only Attention Policy for Long-Form Simultaneous Translation with SpeechLLMs](https://mlsanity.com/paper/2605.31432/) (2026)

## Abstract (arXiv)

We propose a method of segmenting long-form speech by separating semantically complete sentences within the utterance. This prevents the ASR decoder from needlessly processing faraway context while also preventing it from missing relevant context within the current sentence. Semantically complete sentence boundaries are typically demarcated by punctuation in written text; but unfortunately, spoken real-world utterances rarely contain punctuation. We address this limitation by distilling punctuation knowledge from a bidirectional teacher language model (LM) trained on written, punctuated text. We compare our segmenter, which is distilled from the LM teacher, against a segmenter distilled from a acoustic-pause-based teacher used in other works, on a streaming ASR pipeline. The pipeline with our segmenter achieves a 3.2% relative WER gain along with a 60 ms median end-of-segment latency reduction on a YouTube captioning task.