Analysis of unsupervised and noise-robust speaker-adaptive HMM-based speech synthesis systems toward a unified ASR and TTS framework [PDF]
, 2009 For the 2009 Blizzard Challenge we have built an unsupervised version of the HTS-2008 speaker-adaptive HMM-based speech synthesis system for English, and a noise robust version of the systems for Mandarin.
King, S.; id_orcid +13 more
core
Effects of Zeolite and Vermicompost on Changes of Zn Chemical Fractionation in a Polluted Soil
مجله آب و خاک, 2017 Introduction: Soil contamination by heavy metals is a major concern throughout the world, due to persistence of metals in the environment and their toxicity and threat to all living organisms.
Mohsen Hamidpour +3 more
doaj +1 more source
Using Automatic Speech Recognition to Optimize Hearing-Aid Time Constants
Frontiers in Neuroscience, 2022 Automatic speech recognition (ASR), when combined with hearing-aid (HA) and hearing-loss (HL) simulations, can predict aided speech-identification performances of persons with age-related hearing loss.
Lionel Fontan +4 more
doaj +1 more source
Evaluating the Potential Utility of ASR N-Best Lists for Incremental Spoken Dialogue Systems [PDF]
, 2009 Baumann T, Buß O, Schlangen D. Evaluating the Potential Utility of ASR N-Best Lists for Incremental Spoken Dialogue Systems. In: Proceedings of Interspeech 2009.
Schlangen, David ; https://orcid.org/ +2 more
core +1 more source
ASR-Free Pronunciation Assessment [PDF]
Interspeech 2020, 2020 Most of the pronunciation assessment methods are based on local features derived from automatic speech recognition (ASR), e.g., the Goodness of Pronunciation (GOP) score. In this paper, we investigate an ASR-free scoring approach that is derived from the marginal distribution of raw speech signals. The hypothesis is that even if we have no knowledge of
Sitong Cheng +5 more
openaire +2 more sources
The UEDIN English ASR System for the IWSLT 2013 Evaluation [PDF]
, 2013 This paper describes the University of Edinburgh (UEDIN) English ASR system for the IWSLT 2013 Evaluation. Notable features of the system include deep neural network acoustic models in both tandem and hybrid configuration, cross-domain adaptation with ...
Renals, Steve +5 more
core
Durability of Mortar and Concrete with Composite Cements Containing Novel SCMs
Nordic Concrete ResearchThe simplest way to reduce global CO2 emissions linked to cement productions is to reduce the amount of clinker by partially replacing it with supplementary cementitious materials (SCMs). The availability of the most common used SCMs (e.g.
Danner Tobias +5 more
doaj +1 more source
Big Data and quality data for fake news and misinformation detection
Big Data & Society, 2019 Fake news has become an important topic of research in a variety of disciplines including linguistics and computer science. In this paper, we explain how the problem is approached from the perspective of natural language processing, with the goal of ...
Fatemeh Torabi Asr, Maite Taboada
doaj +1 more source
Adapting an Unadaptable ASR System
INTERSPEECH 2023, 2023 As speech recognition model sizes and training data requirements grow, it is increasingly common for systems to only be available via APIs from online service providers rather than having direct access to models themselves. In this scenario it is challenging to adapt systems to a specific target domain.
Ma, Rao +3 more
openaire +3 more sources
sj-pdf-1-asr-10.1177_00031224231184576 – Supplemental material for Pathways of Global Cultural Diffusion: Mass Media and People’s Moral Declarations about Men’s Violence against Women [PDF]
, 2023 Supplemental material, sj-pdf-1-asr-10.1177_00031224231184576 for Pathways of Global Cultural Diffusion: Mass Media and People’s Moral Declarations about Men’s Violence against Women by Jeffrey Swindle in American Sociological ...
Jeffrey Swindle (16700345)
core +1 more source