Joint Disfluency Detection and Constituency Parsing
A joint disfluency detection and constituency parsing model for transcribed speech based on Neural Constituency Parsing of Speech Transcripts from NAACL 2019, with additional changes (e.g. self-training and ensembling) as described in Improving Disfluency Detection by Self-Training a Self-Attentive Model from ACL 2020.
This repository updated the original repository to focus on inferencing using the pretrained swbd_fisher_bert_Edev.0.9078.pt model.
Installation
$ pip install disfluency-constituency-parser
Usage
$ wget https://github.com/pariajm/joint-disfluency-detector-and-parser/releases/download/naacl2019/swbd_fisher_bert_Edev.0.9078.pt
$ wget https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt
$ wget https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased.tar.gz
from dc_parser import DC_Model
model = DC_Model(model_path = "/path/to/swbd_fisher_bert_Edev.0.9078.pt",
bert_model_path = "/path/to/bert-base-uncased.tar.gz",
bert_vocab_path = "/path/to/bert-base-uncased-vocab.txt",)
model.parse(["Today is a very good day!"])
Citation
If you use this model, please cite the following papers:
@inproceedings{jamshid-lou-2019-neural,
title = "Neural Constituency Parsing of Speech Transcripts",
author = "Jamshid Lou, Paria and Wang, Yufei and Johnson, Mark",
booktitle = "Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)",
month = "June",
year = "2019",
address = "Minneapolis, Minnesota",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/N19-1282",
doi = "10.18653/v1/N19-1282",
pages = "2756--2765"
}
@inproceedings{jamshid-lou-2020-improving,
title = "Improving Disfluency Detection by Self-Training a Self-Attentive Model",
author = "Jamshid Lou, Paria and Johnson, Mark",
booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
month = "jul",
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.acl-main.346",
pages = "3754--3763"
}
Release files for disfluency-constituency-parser 0.0.12
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distributions (wheels)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| disfluency_constituency_parser-0.0.12-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl | CPython 3.11 | CPython 3.11 | Linux glibc 2.5+ x86-64, Linux glibc 2.17+ x86-64 | Details |
| disfluency_constituency_parser-0.0.12-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl | CPython 3.10 | CPython 3.10 | Linux glibc 2.17+ x86-64, Linux glibc 2.5+ x86-64 | Details |
| disfluency_constituency_parser-0.0.12-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl | CPython 3.9 | CPython 3.9 | Linux glibc 2.5+ x86-64, Linux glibc 2.17+ x86-64 | Details |
Total release size: 552.4 kB
Release files / disfluency_constituency_parser-0.0.12-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl
| Download URL | disfluency_constituency_parser-0.0.12-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl |
|---|---|
| Size | 184.6 kB |
| Tags | CPython 3.11 Linux glibc 2.17+ x86-64 Linux glibc 2.5+ x86-64 |
|
SHA-256 checksum How to use checksums |
307ba28815ee5b98cd0eee99a8a42391ca26ccaeb7d211c121a45ac568090037
|
|
BLAKE2b-256 checksum How to use checksums |
63e93998ab2e56ce6706328a15f857f63194c52a8aa778c6210c8805f5386fa8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.3
|
Release files / disfluency_constituency_parser-0.0.12-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl
| Download URL | disfluency_constituency_parser-0.0.12-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl |
|---|---|
| Size | 180.9 kB |
| Tags | CPython 3.10 Linux glibc 2.17+ x86-64 Linux glibc 2.5+ x86-64 |
|
SHA-256 checksum How to use checksums |
4e40bd18dd6e25c2dedfe26d8988c33111a38b224f97e1b9786bae9f1226c046
|
|
BLAKE2b-256 checksum How to use checksums |
fd5ef6103e883ccd48afd892c8efd390bb0008d4736989f659d035a9444a0d18
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.3
|
Release files / disfluency_constituency_parser-0.0.12-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl
| Download URL | disfluency_constituency_parser-0.0.12-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl |
|---|---|
| Size | 186.9 kB |
| Tags | CPython 3.9 Linux glibc 2.17+ x86-64 Linux glibc 2.5+ x86-64 |
|
SHA-256 checksum How to use checksums |
7c939d7b409071ad40e13e69454af0b7b259e5dbb42da3aec6dfc3ee6f8f5e7b
|
|
BLAKE2b-256 checksum How to use checksums |
5799509b1ca21fde749dc23ae23df8540e15fdff111de5080f922ba3f02c3a2b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.3
|