# Punctuation Prediction Model for Conversational Speech

arXiv: https://arxiv.org/abs/1807.00543 · HTML: https://mlsanity.com/paper/1807.00543/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Jul 2018.

## Built on by (most cited first, 20 of 20)

- [LibriSpeech-PC: Benchmark for Evaluation of Punctuation and Capitalization Capabilities of end-to-end ASR Models](https://mlsanity.com/paper/2310.02943/) (2023)
- [Token-Level Supervised Contrastive Learning for Punctuation Restoration](https://mlsanity.com/paper/2107.09099/) (2021)
- [A Review of Bangla Natural Language Processing Tasks and the Utility of Transformer Models](https://mlsanity.com/paper/2107.03844/) (2021)
- [Capitalization and Punctuation Restoration: a Survey](https://mlsanity.com/paper/2111.10746/) (2021)
- [Automatic punctuation restoration with BERT models](https://mlsanity.com/paper/2101.07343/) (2021)
- [Improving Punctuation Restoration for Speech Transcripts via External Data](https://mlsanity.com/paper/2110.00560/) (2021)
- [Discriminative Self-training for Punctuation Prediction](https://mlsanity.com/paper/2104.10339/) (2021)
- [Joint prediction of truecasing and punctuation for conversational speech in low-resource scenarios](https://mlsanity.com/paper/2109.06103/) (2021)
- [Unified Multimodal Punctuation Restoration Framework for Mixed-Modality Corpus](https://mlsanity.com/paper/2202.00468/) (2022)
- [Streaming Punctuation for Long-form Dictation with Transformers](https://mlsanity.com/paper/2210.05756/) (2022)
- [Evaluating OpenAI's Whisper ASR for Punctuation Prediction and Topic Modeling of life histories of the Museum of the Person](https://mlsanity.com/paper/2305.14580/) (2023)
- [Longer is (Not Necessarily) Stronger: Punctuated Long-Sequence Training for Enhanced Speech Recognition and Translation](https://mlsanity.com/paper/2409.05601/) (2024)
- [A Context-Aware Feature Fusion Framework for Punctuation Restoration](https://mlsanity.com/paper/2203.12487/) (2022)
- [Streaming Punctuation: A Novel Punctuation Technique Leveraging Bidirectional Context for Continuous Speech Recognition](https://mlsanity.com/paper/2301.03819/) (2023)
- [Multimodal Punctuation Prediction with Contextual Dropout](https://mlsanity.com/paper/2102.11012/) (2021)
- [ViraPart: A Text Refinement Framework for Automatic Speech Recognition and Natural Language Processing Tasks in Persian](https://mlsanity.com/paper/2110.09086/) (2021)
- [Vietnamese Capitalization and Punctuation Recovery Models](https://mlsanity.com/paper/2207.01312/) (2022)
- [PersianPunc: A Large-Scale Dataset and BERT-Based Approach for Persian Punctuation Restoration](https://mlsanity.com/paper/2603.05314/) (2026)
- [A light-weight and efficient punctuation and word casing prediction model for on-device streaming ASR](https://mlsanity.com/paper/2407.13142/) (2024)
- [Restoring Rhythm: Punctuation Restoration Using Transformer Models for Bangla, A Low-Resource Language](https://mlsanity.com/paper/2507.18448/) (2025)

## Abstract (arXiv)

An ASR system usually does not predict any punctuation or capitalization. Lack of punctuation causes problems in result presentation and confuses both the human reader andoff-the-shelf natural language processing algorithms. To overcome these limitations, we train two variants of Deep Neural Network (DNN) sequence labelling models - a Bidirectional Long Short-Term Memory (BLSTM) and a Convolutional Neural Network (CNN), to predict the punctuation. The models are trained on the Fisher corpus which includes punctuation annotation. In our experiments, we combine time-aligned and punctuated Fisher corpus transcripts using a sequence alignment algorithm. The neural networks are trained on Common Web Crawl GloVe embedding of the words in Fisher transcripts aligned with conversation side indicators and word time infomation. The CNNs yield a better precision and BLSTMs tend to have better recall. While BLSTMs make fewer mistakes overall, the punctuation predicted by the CNN is more accurate - especially in the case of question marks. Our results constitute significant evidence that the distribution of words in time, as well as pre-trained embeddings, can be useful in the punctuation prediction task.