Large-Scale Representation Learning from Visually Grounded Untranscribed Speech

Gabriel Ilharco, Yuan Zhang, Jason Baldridge


Abstract
Systems that can associate images with their spoken audio captions are an important step towards visually grounded language learning. We describe a scalable method to automatically generate diverse audio for image captioning datasets. This supports pretraining deep networks for encoding both audio and images, which we do via a dual encoder that learns to align latent representations from both modalities. We show that a masked margin softmax loss for such models is superior to the standard triplet loss. We fine-tune these models on the Flickr8k Audio Captions Corpus and obtain state-of-the-art results—improving recall in the top 10 from 29.6% to 49.5%. We also obtain human ratings on retrieval outputs to better assess the impact of incidentally matching image-caption pairs that were not associated in the data, finding that automatic evaluation substantially underestimates the quality of the retrieved results.
Anthology ID:
K19-1006
Volume:
Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)
Month:
November
Year:
2019
Address:
Hong Kong, China
Venue:
CoNLL
SIG:
SIGNLL
Publisher:
Association for Computational Linguistics
Note:
Pages:
55–65
Language:
URL:
https://www.aclweb.org/anthology/K19-1006
DOI:
10.18653/v1/K19-1006
Bib Export formats:
BibTeX MODS XML EndNote
PDF:
http://aclanthology.lst.uni-saarland.de/K19-1006.pdf