Automated Audio Captioning

Авторы: Кузьмин Никита Валерьевич, Дьяконов Александр Геннадьевич
Сборник: IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events
Год издания: 2020
Место издания: DCASE2020 http://dcase.community/documents/challenge2020/technical_reports/DCASE2020_Kuzmin_137_t6.pdf
Аннотация: This task can be stated as an automated generation textual contentdescription from the raw audio file. We propose a method for theautomated audio captioning task. We examined the impact of augmentations (MixUp, Reverb, Pitch, Over-drive, Speed) on methodperformance. Our method based on modified encoder-decoder architecture. The encoder consists of three bidirectional gated recurrent units (GRU). The decoder consists of one gated recurrent unit(GRU) and one fully-connected layer for classification. The encoder input is log-mel spectrogram features for every part of audiofile segmented by Hann window [1] of 1024 samples with a 50%overlap. The decoder output is a matrix with probabilities of wordsfor each position in a sentence. We used BLEU1, BLEU2, BLEU3,BLEU4, ROUGEL, METEOR, CIDEr, SPICE, SPIDEr metrics tocompare methods.
Добавил в систему: Кузьмин Никита Валерьевич

	ИСТИНА	Войти в систему Регистрация
	ФНКЦ РР
	Главная Поиск Статистика О проекте Помощь

ИСТИНА