TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Audio captioning	Clotho	Ensemble	CIDEr	0.319	# 6
Audio captioning	Clotho	Ensemble	SPIDEr	0.207	# 4
Audio captioning	Clotho	Ensemble	SPICE	0.094	# 4

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/the-ntt-dcase2020-challenge-task-6-system/audio-captioning-on-clotho)](https://paperswithcode.com/sota/audio-captioning-on-clotho?p=the-ntt-dcase2020-challenge-task-6-system)`

The NTT DCASE2020 Challenge Task 6 system: Automated Audio Captioning with Keywords and Sentence Length Estimation

1 Jul 2020 · Yuma Koizumi, Daiki Takeuchi, Yasunori Ohishi, Noboru Harada, Kunio Kashino ·

This technical report describes the system participating to the Detection and Classification of Acoustic Scenes and Events (DCASE) 2020 Challenge, Task 6: automated audio captioning. Our submission focuses on solving two indeterminacy problems in automated audio captioning: word selection indeterminacy and sentence length indeterminacy. We simultaneously solve the main caption generation and sub indeterminacy problems by estimating keywords and sentence length through multi-task learning. We tested a simplified model of our submission using the development-testing dataset. Our model achieved 20.7 SPIDEr score where that of the baseline system was 5.4.

PDF Abstract