Results 1 to 10 of about 29,155,043 (365)

WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing [PDF]

open access: yesIEEE Journal on Selected Topics in Signal Processing, 2021
Self-supervised learning (SSL) achieves great success in speech recognition, while limited exploration has been attempted for other speech processing tasks.
Sanyuan Chen   +16 more
semanticscholar   +1 more source

Is ChatGPT a General-Purpose Natural Language Processing Task Solver? [PDF]

open access: yesConference on Empirical Methods in Natural Language Processing, 2023
Spurred by advancements in scale, large language models (LLMs) have demonstrated the ability to perform a variety of natural language processing (NLP) tasks zero-shot -- i.e., without adaptation on downstream data.
Chengwei Qin   +5 more
semanticscholar   +1 more source

Chemical and Sensory Characteristics of Fruit Juice and Fruit Fermented Beverages and Their Consumer Acceptance

open access: yesBeverages, 2022
Recent social, economic, and technological evolutions have impacted consumption habits. The new consumer is more rational, more connected and demanding with products, more concerned with the management of the family budget, with the health, origin, and ...
Teresa Pinto   +2 more
doaj   +1 more source

Syntactic Comprehension in Persian-Speaking Agrammatics: Further Proof of Trace Deletion Hypothesis [PDF]

open access: yesJournal of Language Horizons, 2021
So far, there have been three important theories explaining the nature and extent of grammatical deficit in agrammatic aphasia. In this respect, while the two mainly performance-based theories of agrammatics, i.e.
Omid Azad
doaj   +1 more source

POTENTIAL OF Entada africana Guill. & Perr. SEED MEAL AS FEED INGREDIENT IN THE DIET OF BROILER CHICKENS

open access: yesTropical and Subtropical Agroecosystems, 2021
Background. The demand and cost of feed ingredients for broiler chicken production have increased in many African countries. This has necessitated the search for alternative feed sources. Objective.
O.J. Makinde   +4 more
doaj   +1 more source

Current status of synthesis and consolidation strategies for thermo-resistant nanoalloys and their general applications

open access: yesNanotechnology Reviews, 2023
Thermo-resistant nanoalloys are a new class of materials that combine high-temperature refractory compounds (such as carbides, nitrides, borides, and oxides) with nanoscale particles of metals, ceramics, or carbon.
Dhand Vivek   +5 more
doaj   +1 more source

COCOA PRODUCTION PATTERN IN NIGERIA: THE MISSING LINK IN REGIONAL AGRO-ECONOMIC DEVELOPMENT [PDF]

open access: yesAnalele Universităţii din Oradea: Seria Geografie, 2020
Pre-independent in Nigeria, cocoa was one of the main source of revenue but lost its glory to the discovery, exploration and exploitation of crude oil, especially from the 70s onwards. Despite the rapid growth in its production and positive impact in the
Olusola Samuel AFOLAYAN
doaj   +1 more source

In-datacenter performance analysis of a tensor processing unit [PDF]

open access: yesInternational Symposium on Computer Architecture, 2017
Many architects believe that major improvements in cost-energy-performance must now come from domain-specific hardware. This paper evaluates a custom ASIC-called a Tensor Processing Unit (TPU)-deployed in datacenters since 2015 that accelerates the ...
Norman P. Jouppi   +75 more
semanticscholar   +1 more source

SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing [PDF]

open access: yesConference on Empirical Methods in Natural Language Processing, 2018
This paper describes SentencePiece, a language-independent subword tokenizer and detokenizer designed for Neural-based text processing, including Neural Machine Translation. It provides open-source C++ and Python implementations for subword units.
Taku Kudo, John Richardson
semanticscholar   +1 more source

Efficient Processing of Deep Neural Networks: A Tutorial and Survey [PDF]

open access: yesProceedings of the IEEE, 2017
Deep neural networks (DNNs) are currently widely used for many artificial intelligence (AI) applications including computer vision, speech recognition, and robotics.
V. Sze   +3 more
semanticscholar   +1 more source

Home - About - Disclaimer - Privacy