We introduce in this work an efficient approach for audio scene
classification using deep recurrent neural networks. An audio scene is firstly
transformed into a sequence of high-level label tree embedding feature vectors.
The vector sequence is then divided into multiple subsequences on which a deep
GRU-based recurrent neural network is trained for sequence-to-label
classification. The global predicted label for the entire sequence is finally
obtained via aggregation of subsequence classification outputs. We will show
that our approach obtains an F1-score of 97.7% on the LITIS Rouen dataset,
which is the largest dataset publicly available for the task. Compared to the
best previously reported result on the dataset, our approach is able to reduce
the relative classification error by 35.3%.

本文使用深度循环神经网络，将音频场景转换为一系列高级标签树嵌入特征向量序列，将其分割为多个子序列，并对序列进行标签分类，最终获得全局预测标签。在 LITIS Rouen 数据集上，实现了 97.7% 的 F1 分数，与该数据集上最佳结果相比，相对分类错误率减少 35.3%。