| viXra.org > Artificial Intelligence > 2208 |
 |
 |
| viXra.org > Artificial Intelligence > 2208 |
|
|
[9] viXra:2208.0173 [pdf] submitted on 2022-08-31 03:40:39
Authors: Mojtaba Heydari, Zhiyao Duan
Comments: 5 Pages.
Online beat tracking (OBT) has always been a challenging task. Dueto the inaccessibility of future data and the need to make inferencein real-time. We propose Don’t Look back! (DLB), a novel approachoptimized for efficiency when performing OBT. DLB feeds theactivations of a unidirectional RNN into an enhanced Monte-Carlolocalization model to infer beat positions. Most preexisting OBTmethods either apply some offline approaches to a moving windowcontaining past data to make predictions about future beat positionsor must be primed with past data at startup to initialize. Meanwhile,our proposed method only uses activation of the current time frameto infer beat positions. As such, without waiting at the beginning toreceive a chunk, it provides an immediate beat tracking response,which is critical for many OBT applications. DLB significantlyimproves beat tracking accuracy over state-of-the-art OBT methods,yielding a similar performance to offline methods.
Category: Artificial Intelligence
[8] viXra:2208.0171 [pdf] submitted on 2022-08-31 03:49:55
Authors: Mojtaba Heydari, Zhiyao Duan
Comments: 8 Pages. 23rd International Society for Music Information Retrieval Conference (ISMIR 2022)
Tracking beats of singing voices without the presence of musical accompaniment can find many applications in music production, automatic song arrangement, and social media interaction.Its main challenge is the lack of strong rhythmic and harmonic patterns that are important for music rhythmic analysis in general. Even for human listeners, this can be a challenging task. As a result, existing music beat tracking systems fail to deliver satisfactory performance on singing voices. In this paper, we propose singing beat tracking as a novel task, and propose the first approach to solving this task. Our approach leverages semantic information of singing voices by employing pre-trained self-supervised WavLM and DistilHuBERT speech representations as the front-end and uses a self-attention encoder layer to predict beats. To train and test the system, we obtain separated singing voices and their beat annotations using source separation and beat tracking on complete songs, followed by manual corrections. Experiments on the 741 separated vocal tracks of the GTZAN dataset show that the proposed system outperforms several state-of-the-art music beat tracking methods by a large margin in terms of beat tracking accuracy. Ablation studies also confirm the advantages of pre-trained self-supervised speech representations over generic spectral features.
Category: Artificial Intelligence
[7] viXra:2208.0156 [pdf] submitted on 2022-08-28 08:46:18
Authors: Carlo D. Petalver
Comments: 12 Pages.
Categorizing books and other archaic paper sources to a course reference or syllabus is a challenge in library science. The traditional way of categorization is manually done by professionals and the process of seeking and retrieving information can be frustrating. It needs intellectual tasks and conceptual analysis of a human effort to recognize similarities of items in determining the subject to the correct category. Unlike the traditional categorization process, the author implemented the concept of automatic document categorization for libraries using text mining. The project involves the creation of a web app and mobile app. This can be accomplished through the use of a supervised machine learning classification model using the Support Vector Machine algorithm that can predict the given category of data from the book or other archaic paper sources to the course syllabus they belong to.
Category: Artificial Intelligence
[6] viXra:2208.0137 [pdf] submitted on 2022-08-25 15:44:36
Authors: Yingcheng Huang, Fuyuan Xiao
Comments: 1 Page.
In this paper, a novel belief divergence, higher order belief Jensen-Shannon divergence is proposedto measure the discrepancy between BPAs in Dempster—Shafer evidence theory.
Category: Artificial Intelligence
[5] viXra:2208.0135 [pdf] submitted on 2022-08-25 00:53:10
Authors: Jie Zenga, Fuyuan Xiao
Comments: Pages.
In this paper, a novel symmetric fractal-based belief KL divergence is proposed to more appropriately measure the conflict between BPAs.
Category: Artificial Intelligence
[4] viXra:2208.0104 [pdf] submitted on 2022-08-20 05:18:24
Authors: Akhil Sahukaru, Shishir Kumar Shandiliya
Comments: 15 Pages.
When traffic demand exceeds available network capacity, traffic congestion develops. Lower vehicle speeds, longer journey times, unreliable arrival timings, and lengthiervehicular queueing are all symptoms. Congestion may have a detrimental influence on society bylowering quality of life and increasing pollution, particularly in metropolitan areas. To alleviatetraffic congestion, traffic engineers and scientists require high-quality, comprehensive, andprecise data to forecast traffic flow. The advantages and disadvantages of various data collectingsystems, as well as data attributes such as accuracy, sample frequency, and geographiccoverage, vary. Multisource data fusion improves accuracy and delivers a more complete picture of trafficflow performance on a road network. This study provides a review of the literature on congestionestimation and prediction based on data obtained from numerous sources. An overview of datafusion approaches and congestion indicators that have been employed in the literature to estimatetraffic condition and congestion is provided. The outcomes of various strategies are examined,and a disseminative analysis of the benefits and drawbacks of the methods reviewed is offered.Keywords: traffic congestion; multi source data fusion; traffic state estimation; data collection
Category: Artificial Intelligence
[3] viXra:2208.0073 [pdf] submitted on 2022-08-13 01:00:59
Authors: Mirzakhmet Syzdykov
Comments: 3 Pages.
We propose the evolutionary algorithm for subset construction which superceeds previous known resultdue to Rabin and Scott.
Category: Artificial Intelligence
[2] viXra:2208.0055 [pdf] submitted on 2022-08-09 13:40:27
Authors: Egger L Mielberg
Comments: 17 Pages.
Time is the most important asset of any living person on our planet.The presence of a digital personal financial and economic environment, decentralized to each of its users, would significantly change the quality and standard of living of this user.The main unit of measurement of the value of an individual user of the environment should be the hours (minutes) spent by him on the execution of any sense contract.Our international team proposes a practical implementation of such an environment using the logic of the new mathematical theory for artificial intelligence Sense Theory [1].
Category: Artificial Intelligence
[1] viXra:2208.0012 [pdf] submitted on 2022-08-04 01:28:39
Authors: Michael C. I. Nwogugu
Comments: 32 Pages. The copyright license-type for this article is CC-BY-NC-ND
Nwogugu (2012) introduced a Network-based and Cognition-Based cyberphysical fuzzy-system within which complex self-adjusting "semi-autonomous" financial products are originated, purchased and sold. The participants of the system are diverse and include adults, companies, brokers, banks, lawyers, insurance companies and real estate companies. This theoretical article explains the key additional characteristics, system-architecture, fuzzy-attributes and Reasoning/Logic of some cost-reducing and energy-reducing AI/ML Network/Modular Products (ie. Mortgage-Alternatives Products, Retirement/Savings products and Insurance products) that were introduced in Nwogugu (2012), and also other cost-saving financial products that he developed (collectively, the "Products"). Through the products’ fuzzy features, AI and network, the cyber-system architecture implicitly incorporates "Learning" and also can use Blockchain for record-keeping. The semi-autonomous and "self-adjustment" characteristics of these Modular Products can drastically reduce system-participants’ costs and energy-use while increasing their revenues/profits through better and more efficient CRM, "matching", transaction-processing and "state-updating".
Category: Artificial Intelligence
| Third-Party Links: |
|