Satoru Fukayama

Shikhar Bharadwaj, Samuele Cornell, Kwanghee Choi et al. · cmu

Masked token prediction has emerged as a powerful pre-training objective across language, vision, and speech, offering the potential to unify these diverse modalities through a single pre-training task. However, its application for general audio understanding remains underexplored, with BEATs being the only notable example. BEATs has seen limited modifications due to the absence of open-source pre-training code. Furthermore, BEATs was trained only on AudioSet, restricting its broader downstream applicability. To address these gaps, we present OpenBEATs, an open-source framework that extends BEATs via multi-domain audio pre-training. We conduct comprehensive evaluations across six types of tasks, twenty five datasets, and three audio domains, including audio reasoning tasks such as audio question answering, entailment, and captioning. OpenBEATs achieves state-of-the-art performance on six bioacoustics datasets, two environmental sound datasets and five reasoning datasets, performing better than models exceeding a billion parameters at one-fourth their parameter size. These results demonstrate the effectiveness of multi-domain datasets and masked token prediction task to learn general-purpose audio representations. To promote further research and reproducibility, we release all pre-training and evaluation code, pretrained and fine-tuned checkpoints, and training logs at https://github.com/Shikhar-S/OpenBEATs

Tomohiko Nakamura, Shinnosuke Takamichi, Naoko Tanji et al.

We construct a corpus of Japanese a cappella vocal ensembles (jaCappella corpus) for vocal ensemble separation and synthesis. It consists of 35 copyright-cleared vocal ensemble songs and their audio recordings of individual voice parts. These songs were arranged from out-of-copyright Japanese children's songs and have six voice parts (lead vocal, soprano, alto, tenor, bass, and vocal percussion). They are divided into seven subsets, each of which features typical characteristics of a music genre such as jazz and enka. The variety in genre and voice part match vocal ensembles recently widespread in social media services such as YouTube, although the main targets of conventional vocal ensemble datasets are choral singing made up of soprano, alto, tenor, and bass. Experimental evaluation demonstrates that our corpus is a challenging resource for vocal ensemble separation. Our corpus is available on our project page (https://tomohikonakamura.github.io/jaCappella_corpus/).

Kwanghee Choi, Ankita Pasad, Tomohiko Nakamura et al.

Self-supervised speech models (S3Ms) have become an effective backbone for speech applications. Various analyses suggest that S3Ms encode linguistic properties. In this work, we seek a more fine-grained analysis of the word-level linguistic properties encoded in S3Ms. Specifically, we curate a novel dataset of near homophone (phonetically similar) and synonym (semantically similar) word pairs and measure the similarities between S3M word representation pairs. Our study reveals that S3M representations consistently and significantly exhibit more phonetic than semantic similarity. Further, we question whether widely used intent classification datasets such as Fluent Speech Commands and Snips Smartlights are adequate for measuring semantic abilities. Our simple baseline, using only the word identity, surpasses S3M-based models. This corroborates our findings and suggests that high scores on these datasets do not necessarily guarantee the presence of semantic content.

Takayuki Nakatsuka, Kazuyoshi Yoshii, Yuki Koyama et al.

This paper proposes a statistical approach to 2D pose estimation from human images. The main problems with the standard supervised approach, which is based on a deep recognition (image-to-pose) model, are that it often yields anatomically implausible poses, and its performance is limited by the amount of paired data. To solve these problems, we propose a semi-supervised method that can make effective use of images with and without pose annotations. Specifically, we formulate a hierarchical generative model of poses and images by integrating a deep generative model of poses from pose features with that of images from poses and image features. We then introduce a deep recognition model that infers poses from images. Given images as observed data, these models can be trained jointly in a hierarchical variational autoencoding (image-to-pose-to-feature-to-pose-to-image) manner. The results of experiments show that the proposed reflective architecture makes estimated poses anatomically plausible, and the performance of pose estimation improved by integrating the recognition and generative models and also by feeding non-annotated images.

Yin-Cheng Yeh, Wen-Yi Hsiao, Satoru Fukayama et al.

Several prior works have proposed various methods for the task of automatic melody harmonization, in which a model aims to generate a sequence of chords to serve as the harmonic accompaniment of a given multiple-bar melody sequence. In this paper, we present a comparative study evaluating and comparing the performance of a set of canonical approaches to this task, including a template matching based model, a hidden Markov based model, a genetic algorithm based model, and two deep learning based models. The evaluation is conducted on a dataset of 9,226 melody/chord pairs we newly collect for this study, considering up to 48 triad chords, using a standardized training/test split. We report the result of an objective evaluation using six different metrics and a subjective study with 202 participants.