https://github.com/kaituoxu/X-Punctuator

A PyTorch implementation of a punctuation prediction system using (B)LSTM, which automatically adds suitable punctuation into text without punctuation.
https://github.com/kaituoxu/X-Punctuator

lstm punctuation pytorch

Last synced: 10 days ago
JSON representation

A PyTorch implementation of a punctuation prediction system using (B)LSTM, which automatically adds suitable punctuation into text without punctuation.

• Host: GitHub
• URL: https://github.com/kaituoxu/X-Punctuator
• Owner: kaituoxu
• Created: 2019-02-10T11:54:56.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2020-05-13T09:18:21.000Z (about 4 years ago)
• Last Synced: 2024-03-05T12:38:28.615Z (4 months ago)
• Topics: lstm, punctuation, pytorch
• Language: Python
• Homepage:
• Size: 158 KB
• Stars: 60
• Watchers: 3
• Forks: 21
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Lists

• awesome-punctuator - X-*Punctuator*

README

        
# X-Punctuator

A PyTorch implementation of punctuation prediction system using LSTM/BLSTM [1][2][3], which automatically adds suitable punctation into text without punctuation.

## Install

- PyTorch 0.4+

## Usage

`egs/toy/run.sh` provides an example usage.

```bash

# Set PATH and PYTHONPATH

$ cd egs/toy/; . ./path.sh

# Train

$ train.py -h

# Add punctuation

$ add_punctuation.py -h

# Analyze metrics

$ analyer.py -h

```

#### How to visualize loss?

If you want to visualize your loss, you can use [visdom](https://github.com/facebookresearch/visdom) to do that:

1. Open a new terminal in your remote server (recommend tmux) and run `$ visdom`.

2. Open a new terminal and run `$ train.py ... --visdom 1 --vidsdom_id ""`.

3. Open your browser and type `:8097`, egs, `127.0.0.1:8097`.

4. In visdom website, chose `` in `Environment` to see your loss.

#### How to resume training?

```bash

$ train.py --continue_from 

```

#### How to use multi-GPU?

Use comma separated gpu-id sequence, such as:

```bash

$ CUDA_VISIBLE_DEVICES="0,1" train.py

```

## Reference

- [1] Kaituo Xu, Lei Xie, and Kaisheng Yao. "Investigating LSTM for punctuation prediction" in ISCSLP 2016

- [2] Ottokar Tilk and Tanel Alumae. "Bidirectional Recurrent Neural Network with Attention Mechanism for Punctuation Restoration" in Interspeech 2016

- [3] Ottokar Tilk and Tanel Alumae. "LSTM for Punctuation Restoration in Speech Transcripts" in Interspeech 2015