Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech

ResDAVEnet-VQ

Official PyTorch implementation of Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech

What is in this repo?

• Multi-GPU training of ResDAVEnet-VQ
• Quantitative evaluation
  • Image-to-speech and speech-to-image retrieval
  • ZeroSpeech 2019 ABX phone-discriminability test
  • Word detection
• Qualitative evaluation
  • Visualize time-aligned word/phone/code transcripts
  • F1/recall/precision scatter plots for model/layer comparison

If you find the code useful, please cite

@inproceedings{Harwath2020Learning,
  title={Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech},
  author={David Harwath and Wei-Ning Hsu and James Glass},
  booktitle={International Conference on Learning Representations},
  year={2020},
  url={https://openreview.net/forum?id=B1elCp4KwH}
}

Pre-trained models

Data preparation

Download the MIT Places Image/Audio Data

We use MIT Places scene recognition database (Places Image) and a paired
MIT Places Audio Caption Corpus (Places Audio) as visually-grounded speech,
which contains roughly 400K image/spoken caption pairs, to train ResDAVEnet-VQ.

• Places Image can be downloaded here
• Places Audio can be downloaded here

Optional data preprocessing

Data specifcation files can be found at metadata/{train,val}.json inside the
Places Audio directory; however, they do not include the time-aligned word
transcripts for analysis. Those with alignments can be downloaded here:

• train
• valid

Open the *.json files and update the values of image_base_path and
audio_base_path to reflect the path where the image and the audio datasets
are stored.

To speed up data loading, we save images and audio data into the
HDF5 binary files, and use the
h5py Python interface to access the data.
The corresponding PyTorch Dataset class is ImageCaptionDatasetHDF5 in
./dataloaders/image_caption_dataset_hdf5.py. To prepare HDF5 datasets, run

./scripts/preprocess.sh

(We do support on-the-fly feature processing with the ImageCaptionDataset class
in ./dataloaders/image_caption_dataset.py, which takes a data specification
file as input (e.g., metadata/train.json). However, this can be very slow)

ImageCaptionDataset and ImageCaptionDatasetHDF5 are interchangeable, but
most scripts in this repo assume the preprocessed HDF5 dataset is available.
Users would have to modify the code correspondingly to use
ImageCaptionDataset.

Interactive Qualtitative Evaluation

See run_evaluations.ipynb

Quantitative Evaluation

ZeroSpeech 2019 ABX Phone Discriminability Test

Users need to download the dataset and the Docker image by following the
instructions here.

To extract ResDAVEnet-VQ features, see ./scripts/dump_zs19_abx.sh.

Word detection

See ./run_unit_analysis.py. It needs both HDF5 dataset and the original JSON
dataset to get the time-aligned word transcripts.

Example:

python run_unit_analysis.py --hdf5_path=$hdf5_path --json_path=$json_path \
  --exp_dir=$exp_dir --layer=$layer --output_dir=$out_dir

Cross-modal retrieval

See ./run_ResDavenetVQ.py. Set --mode=eval for retrieval evaluation.

Example:

python run_ResDavenetVQ.py --resume=True --mode=eval \
  --data-train=$data_tr --data-val=$data_dt \
  --exp-dir="./exps/pretrained/RDVQ_01000_01100_01110"

Training

See ./scripts/train.sh.

To train a model from scratch with the 2nd and 3rd layers quantized, run

./scripts/train.sh 01100 RDVQ_01100 ""

To train a model with the 2nd and 3rd layers quantized, and initialize weights
from a pre-trained model (e.g., ./exps/RDVQ_00000), run

./scripts/train.sh 01100 RDVQ_01100 "--seed-dir ./exps/RDVQ_00000"

GitHub

https://github.com/wnhsu/ResDAVEnet-VQ