Voice activity detection in the wild: A data-driven approach using teacher-student training

Voice activity detection is an essential pre-processing component for speech-related tasks such as automatic speech recognition (ASR). Traditional supervised VAD systems obtain frame-level labels from an ASR pipeline by using, e.g., a Hidden Markov model. These ASR models are commonly trained on clean and fully transcribed data, limiting VAD systems to be trained on clean or synthetically noised datasets. Therefore, a major challenge for supervised VAD systems is their generalization towards noisy, real-world data. This work proposes a data-driven teacher-student approach for VAD, which utilizes vast and unconstrained audio data for training. Unlike previous approaches, only weak labels during teacher training are required, enabling the utilization of any real-world, potentially noisy dataset. Our approach firstly trains a teacher model on a source dataset (Audioset) using clip-level supervision. After training, the teacher provides frame-level guidance to a student model on an unlabeled, target dataset. A multitude of student models trained on mid- to large-sized datasets are investigated (Audioset, Voxceleb, NIST SRE). Our approach is then respectively evaluated on clean, artificially noised, and real-world data. We observe significant performance gains in artificially noised and real-world scenarios. Lastly, we compare our approach against other unsupervised and supervised VAD methods, demonstrating our method's superiority.

Text-to-Audio Grounding: Building Correspondence Between Captions and Sound Events

Automated Audio Captioning is a cross-modal task, generating natural language descriptions to summarize the audio clips' sound events. However, grounding the actual sound events in the given audio based on its corresponding caption has not been investigated. This paper contributes an AudioGrounding dataset, which provides the correspondence between sound events and the captions provided in Audiocaps, along with the location (timestamps) of each present sound event. Based on such, we propose the text-to-audio grounding (TAG) task, which interactively considers the relationship between audio processing and language understanding. A baseline approach is provided, resulting in an event-F1 score of 28.3% and a Polyphonic Sound Detection Score (PSDS) score of 14.7%.

Investigating Local and Global Information for Automated Audio Captioning with Transfer Learning

Automated audio captioning (AAC) aims at generating summarizing descriptions for audio clips. Multitudinous concepts are described in an audio caption, ranging from local information such as sound events to global information like acoustic scenery. Currently, the mainstream paradigm for AAC is the end-to-end encoder-decoder architecture, expecting the encoder to learn all levels of concepts embedded in the audio automatically. This paper first proposes a topic model for audio descriptions, comprehensively analyzing the hierarchical audio topics that are commonly covered. We then explore a transfer learning scheme to access local and global information. Two source tasks are identified to respectively represent local and global information, being Audio Tagging (AT) and Acoustic Scene Classification (ASC). Experiments are conducted on the AAC benchmark dataset Clotho and Audiocaps, amounting to a vast increase in all eight metrics with topic transfer learning. Further, it is discovered that local information and abstract representation learning are more crucial to AAC than global information and temporal relationship learning.

Towards duration robust weakly supervised sound event detection

Sound event detection (SED) is the task of tagging the absence or presence of audio events and their corresponding interval within a given audio clip. While SED can be done using supervised machine learning, where training data is fully labeled with access to per event timestamps and duration, our work focuses on weakly-supervised sound event detection (WSSED), where prior knowledge about an event's duration is unavailable. Recent research within the field focuses on improving segment- and event-level localization performance for specific datasets regarding specific evaluation metrics. Specifically, well-performing event-level localization requires fully labeled development subsets to obtain event duration estimates, which significantly benefits localization performance. Moreover, well-performing segment-level localization models output predictions at a coarse-scale (e.g., 1 second), hindering their deployment on datasets containing very short events (< 1 second). This work proposes a duration robust CRNN (CDur) framework, which aims to achieve competitive performance in terms of segment- and event-level localization. This paper proposes a new post-processing strategy named "Triple Threshold" and investigates two data augmentation methods along with a label smoothing method within the scope of WSSED. Evaluation of our model is done on the DCASE2017 and 2018 Task 4 datasets, and URBAN-SED. Our model outperforms other approaches on the DCASE2018 and URBAN-SED datasets without requiring prior duration knowledge. In particular, our model is capable of similar performance to strongly-labeled supervised models on the URBAN-SED dataset. Lastly, ablation experiments to reveal that without post-processing, our model's localization performance drop is significantly lower compared with other approaches.

Multiple Sound Sources Localization from Coarse to Fine

How to visually localize multiple sound sources in unconstrained videos is a formidable problem, especially when lack of the pairwise sound-object annotations. To solve this problem, we develop a two-stage audiovisual learning framework that disentangles audio and visual representations of different categories from complex scenes, then performs cross-modal feature alignment in a coarse-to-fine manner. Our model achieves state-of-the-art results on public dataset of localization, as well as considerable performance on multi-source sound localization in complex scenes. We then employ the localization results for sound separation and obtain comparable performance to existing methods. These outcomes demonstrate our model's ability in effectively aligning sounds with specific visual sources. Code is available at https://github.com/shvdiwnkozbw/Multi-Source-Sound-Localization

What does a Car-ssette tape tell?

Captioning has attracted much attention in image and video understanding while little work examines audio captioning. This paper contributes a manually-annotated dataset on car scene, in extension to a previously published hospital audio captioning dataset. An encoder-decoder model with pretrained word embeddings and additional sentence loss is proposed. This current model can accelerate the training process and generate semantically correct but unseen unique sentences. We test the model on the current car dataset, previous Hospital Dataset and the Joint Dataset, indicating its generalization capability across different scenes. Further, we make an effort to provide a better objective evaluation metric, namely the BERT similarity score. It compares the semantic-level similarity and compensates for drawbacks of N-gram based metrics like BLEU, namely high scores for word-similar sentences. This new metric demonstrates higher correlation with human evaluation. However, though detailed audio captions can now be automatically generated, human annotations still outperform model captions in many aspects.

Text-based Depression Detection: What Triggers An Alert

Recent advances in automatic depression detection mostly derive from modality fusion and deep learning methods. However multi-modal approaches insert significant difficulty in data collection phase while deep learning methods' opaqueness lowers its credibility. This current work proposes a text-based multi-task BLSTM model with pretrained word embeddings. Our method outputs depression presence results as well as predicted severity score, culminating a state-of-the-art F1 score of 0.87, outperforming previous multi-modal studies. We also achieve the lowest RMSE compared with currently available text-based approaches. Further, by utilizing a per time step attention mechanism we analyse the sentences/words that contribute most in predicting the depressed state. Surprisingly, `unmeaningful' words/paralinguistic information such as `um' and `uh' are the indicators to our model when making a depression prediction. It is for the first time revealed that fillers in a conversation trigger a depression alert for a deep learning model.

Audio Caption: Listen and Tell

Increasing amount of research has shed light on machine perception of audio events, most of which concerns detection and classification tasks. However, human-like perception of audio scenes involves not only detecting and classifying audio sounds, but also summarizing the relationship between different audio events. Comparable research such as image caption has been conducted, yet the audio field is still quite barren. This paper introduces a manually-annotated dataset for audio caption. The purpose is to automatically generate natural sentences for audio scene description and to bridge the gap between machine perception of audio and image. The whole dataset is labelled in Mandarin and we also include translated English annotations. A baseline encoder-decoder model is provided for both English and Mandarin. Similar BLEU scores are derived for both languages: our model can generate understandable and data-related captions based on the dataset.