Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845564
Y. Oyedele, Darelle van Greunen
In the human computer interaction domain, there are few to no design guidelines that take into consideration the holistic end-user engagement and interaction with the ICT product. The extract from the study presented in this paper aimed at achieving two objectives: to determine the parts of a process for the development of end-user experience (UX) heuristics, and to identify UX heuristics for engagement and interaction using the heuristics development process. The study employed the use of qualitative data collection and analysis techniques. The primary contributions in this paper are the formulated heuristics development approach, and the UX heuristics for engagement and interaction aligned to the three dimensions of UX engagement and interaction namely emotional, cognitive, and behavioural. Many of the guidelines support at least two UX engagement and interaction dimensions, thus showing that designing for an engaging experience needs to consider not just a single dimension, but a dual or more holistic approach.
{"title":"Towards User Experience Heuristics for Engagement and Interaction","authors":"Y. Oyedele, Darelle van Greunen","doi":"10.23919/IST-Africa56635.2022.9845564","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845564","url":null,"abstract":"In the human computer interaction domain, there are few to no design guidelines that take into consideration the holistic end-user engagement and interaction with the ICT product. The extract from the study presented in this paper aimed at achieving two objectives: to determine the parts of a process for the development of end-user experience (UX) heuristics, and to identify UX heuristics for engagement and interaction using the heuristics development process. The study employed the use of qualitative data collection and analysis techniques. The primary contributions in this paper are the formulated heuristics development approach, and the UX heuristics for engagement and interaction aligned to the three dimensions of UX engagement and interaction namely emotional, cognitive, and behavioural. Many of the guidelines support at least two UX engagement and interaction dimensions, thus showing that designing for an engaging experience needs to consider not just a single dimension, but a dual or more holistic approach.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130767298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845537
Julien Ntamakemwa, J. Njenga
Tech start-ups have various needs, both financial and non-financial. Like financial support, non-financial support needs contribute to the success of tech startups. It is crucial to understand the tech start-ups’ non-financial needs. A qualitative study was conducted with eight digital entrepreneurs (DEs) of tech start-ups in South Africa. The study aimed at investigating the different non-financial support needs tech start-ups have and the non-financial support they have received. Virtual semi-structured interviews were conducted with the DEs to understand the non-financial support they received and wish to receive. Non-financial support needs were identified, including branding, legal, access to market, mentoring, coaching, advisory support, and access to incubators and technology hubs. The legal support was identified as beneficial as it provided tech start-ups with the regulations needed to be followed for their tech start-up. The study also made recommendations.
{"title":"Non-Financial Support Needs of Tech Start-ups to Minimise Risk of Failure","authors":"Julien Ntamakemwa, J. Njenga","doi":"10.23919/IST-Africa56635.2022.9845537","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845537","url":null,"abstract":"Tech start-ups have various needs, both financial and non-financial. Like financial support, non-financial support needs contribute to the success of tech startups. It is crucial to understand the tech start-ups’ non-financial needs. A qualitative study was conducted with eight digital entrepreneurs (DEs) of tech start-ups in South Africa. The study aimed at investigating the different non-financial support needs tech start-ups have and the non-financial support they have received. Virtual semi-structured interviews were conducted with the DEs to understand the non-financial support they received and wish to receive. Non-financial support needs were identified, including branding, legal, access to market, mentoring, coaching, advisory support, and access to incubators and technology hubs. The legal support was identified as beneficial as it provided tech start-ups with the regulations needed to be followed for their tech start-up. The study also made recommendations.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125478367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845603
W. Maina, L. Nderu, Tobias Mwalili
The current practice of informal cyber threat intelligence (CTI) sharing among organizations is characterized with the use of emails and social media exchanges among individuals. This model is highly subjective and dependent on a specific individual’s social networks. Financial institutions have CTI at their disposal that could protect each other from computer hacks and fraud. The effective sharing of this intelligence among financial institutions could reduce the high income leakages that is brought about by cyber-attacks. The challenge is how to share this intelligence confidentially and anonymously since the financial institutions are competitors, have a huge reputation to protect and thrive on business secrecy. This research proposes a new way of sharing cyber threat intelligence by using Ethereum smart contract blockchain technology. This will be accomplished by hashing the device identity and replacing the hashed device identity by an on-chain verifiable random function to protect the identity of the participating nodes or the financial institutions in the blockchain network when passing information.
{"title":"A Smart Contract Approach to Cyber Threat Intelligence Sharing in Kenya","authors":"W. Maina, L. Nderu, Tobias Mwalili","doi":"10.23919/IST-Africa56635.2022.9845603","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845603","url":null,"abstract":"The current practice of informal cyber threat intelligence (CTI) sharing among organizations is characterized with the use of emails and social media exchanges among individuals. This model is highly subjective and dependent on a specific individual’s social networks. Financial institutions have CTI at their disposal that could protect each other from computer hacks and fraud. The effective sharing of this intelligence among financial institutions could reduce the high income leakages that is brought about by cyber-attacks. The challenge is how to share this intelligence confidentially and anonymously since the financial institutions are competitors, have a huge reputation to protect and thrive on business secrecy. This research proposes a new way of sharing cyber threat intelligence by using Ethereum smart contract blockchain technology. This will be accomplished by hashing the device identity and replacing the hashed device identity by an on-chain verifiable random function to protect the identity of the participating nodes or the financial institutions in the blockchain network when passing information.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121610764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845568
Diane Niyomwungere, W. Mwangi, R. Rimiru
Maize is an essential cereal for humans and animals worldwide, and it is one of the staple food in Kenya. One of the main challenges facing the maize crop in Kenya is the presence of diseases spreading quickly. Early recognition of maize pathogen and disease help at preventing the disease from spreading throughout the field. This paper proposes a regularized Multitask learning (MTL)–Convolutional Neural Networks (CNN) model for simultaneously identifying maize disease and its pathogen from diseased maize images. MTL allows training one model for multiple tasks at a time, which may improve the accuracy of each task by taking advantage of their commonalities. Our baseline is made of two CNN classification models, one of them being overfitting. We then build an MTL based on the two models, which increases the test accuracy of the overfitting model from 60.08% to 74.48%. The results show that the accuracy rises to 77.44% while combining MTL to the Early stopping method. However, the test accuracy goes up to 85.22 percent when MTL is combined with Early Stopping and Transfer Learning. The model is deployed to an android mobile application for maize farmers as end-users which is very important for costs reduction and time saving.
{"title":"Multi-task Neural Networks Convolutional Learning Model for Maize Disease Identification","authors":"Diane Niyomwungere, W. Mwangi, R. Rimiru","doi":"10.23919/IST-Africa56635.2022.9845568","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845568","url":null,"abstract":"Maize is an essential cereal for humans and animals worldwide, and it is one of the staple food in Kenya. One of the main challenges facing the maize crop in Kenya is the presence of diseases spreading quickly. Early recognition of maize pathogen and disease help at preventing the disease from spreading throughout the field. This paper proposes a regularized Multitask learning (MTL)–Convolutional Neural Networks (CNN) model for simultaneously identifying maize disease and its pathogen from diseased maize images. MTL allows training one model for multiple tasks at a time, which may improve the accuracy of each task by taking advantage of their commonalities. Our baseline is made of two CNN classification models, one of them being overfitting. We then build an MTL based on the two models, which increases the test accuracy of the overfitting model from 60.08% to 74.48%. The results show that the accuracy rises to 77.44% while combining MTL to the Early stopping method. However, the test accuracy goes up to 85.22 percent when MTL is combined with Early Stopping and Transfer Learning. The model is deployed to an android mobile application for maize farmers as end-users which is very important for costs reduction and time saving.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116275149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845566
Gerald Tsimba, Beauty Mugoniwa, A. Mutembedza
This paper was motivated by the increased digital disparity between rural and urban learners in Zimbabwe. It was further exacerbated by the advent of Covid-19 pandemic in 2020, which disenfranchised rural learners as schools had no other alternative, except adopting eLearning platforms. The study’s main objectives were to determine the preparedness of rural schools in adopting eLearning and conducting a comparative analysis between rural and urban schools on equitable access to eLearning during Covid-19 crisis. In that respect, the study pursued a mixed research methodology grounded on a pragmatist philosophical view. In line with that, an exploratory case study strategy became prime in gathering both qualitive and quantitative from 112 participants found in Goromonzi district schools of Zimbabwe. Ethnography was used to gather qualitative data between 2020 and 2021, whilst a survey contributed to quantitative data gathering. In that respect, findings from the study suggests that rural schools were largely disadvantaged by the adoption of eLearning during the Covid-19 pandemic period, as very few learners could attend online schools. More so, there was an increased educational inequality between rural and urban learners, mainly attributed to lack of ICT infrastructure and resources. Therefore, the study proposed an eLearning adoption strategy, which could be adopted by key stakeholders in the school education system during crisis periods.
{"title":"Equitable Access to eLearning during Covid-19 Pandemic and beyond. A Comparative Analysis between Rural and Urban Schools in Zimbabwe","authors":"Gerald Tsimba, Beauty Mugoniwa, A. Mutembedza","doi":"10.23919/IST-Africa56635.2022.9845566","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845566","url":null,"abstract":"This paper was motivated by the increased digital disparity between rural and urban learners in Zimbabwe. It was further exacerbated by the advent of Covid-19 pandemic in 2020, which disenfranchised rural learners as schools had no other alternative, except adopting eLearning platforms. The study’s main objectives were to determine the preparedness of rural schools in adopting eLearning and conducting a comparative analysis between rural and urban schools on equitable access to eLearning during Covid-19 crisis. In that respect, the study pursued a mixed research methodology grounded on a pragmatist philosophical view. In line with that, an exploratory case study strategy became prime in gathering both qualitive and quantitative from 112 participants found in Goromonzi district schools of Zimbabwe. Ethnography was used to gather qualitative data between 2020 and 2021, whilst a survey contributed to quantitative data gathering. In that respect, findings from the study suggests that rural schools were largely disadvantaged by the adoption of eLearning during the Covid-19 pandemic period, as very few learners could attend online schools. More so, there was an increased educational inequality between rural and urban learners, mainly attributed to lack of ICT infrastructure and resources. Therefore, the study proposed an eLearning adoption strategy, which could be adopted by key stakeholders in the school education system during crisis periods.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115249528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845550
Petrus Helao Haixula, J. Osakwe
Recent studies on the outsourcing of Information Systems (IS) have shown that there is a growing trend and widely accepted management practice of IS outsourcing in the public sector. Therefore, it is important to investigate the effectiveness of outsourcing information Systems in these government agencies and parastatals. A quantitative research design was used for this study, and data was collected from the staff of the Namibian Ministry of Labour. Purposive sampling technique was used to select 25 staff members through the use of survey questionnaires while descriptive statistics was used to analyse data using the Statistical Package for Social Sciences (SPSS). The findings of this study reveal the lack of due process being followed which leads to the failure of the process.
{"title":"Effectiveness of Outsourced Information Systems in Public Sector","authors":"Petrus Helao Haixula, J. Osakwe","doi":"10.23919/IST-Africa56635.2022.9845550","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845550","url":null,"abstract":"Recent studies on the outsourcing of Information Systems (IS) have shown that there is a growing trend and widely accepted management practice of IS outsourcing in the public sector. Therefore, it is important to investigate the effectiveness of outsourcing information Systems in these government agencies and parastatals. A quantitative research design was used for this study, and data was collected from the staff of the Namibian Ministry of Labour. Purposive sampling technique was used to select 25 staff members through the use of survey questionnaires while descriptive statistics was used to analyse data using the Statistical Package for Social Sciences (SPSS). The findings of this study reveal the lack of due process being followed which leads to the failure of the process.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115258800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845674
Sifiso Dlamini, Marita Turpin, M. Herselman
The global spread of corona virus has had a significant impact on the basic education systems across the world. Nation states and various government departments had to invent means for remote schooling particularly to ensure continuation of learning amidst the pandemic. One of the widely used means of teaching and learning adopted across the world was the use of mobile technologies for remote learning and content sharing. This directly tested the teachers and learners’ resilience in using mobile technologies for teaching and learning. A scoping literature review was conducted to identify technology and pedagogical factors which affected the use of technology for teaching and learning during 2020 school closure. The findings of the literature review identify access to mobile technologies, pedagogical factors like effective usage of technology for teaching and learning, and connectivity as the main barriers for schools in rural areas. Findings from this exercise are used as factors to be considered when building a resilience framework for introducing and using mobile technologies in South African rural schools. The purpose of undertaking such a task is to complete a continuous evaluation of the need to develop a resilience framework and guidelines for the introduction and use of mobile technologies in South African rural schools.
{"title":"Technology and Pedagogical Factors to be Considered when Building a Resilience Framework for Integrating and Using Mobile Technologies in South African Rural Schools","authors":"Sifiso Dlamini, Marita Turpin, M. Herselman","doi":"10.23919/IST-Africa56635.2022.9845674","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845674","url":null,"abstract":"The global spread of corona virus has had a significant impact on the basic education systems across the world. Nation states and various government departments had to invent means for remote schooling particularly to ensure continuation of learning amidst the pandemic. One of the widely used means of teaching and learning adopted across the world was the use of mobile technologies for remote learning and content sharing. This directly tested the teachers and learners’ resilience in using mobile technologies for teaching and learning. A scoping literature review was conducted to identify technology and pedagogical factors which affected the use of technology for teaching and learning during 2020 school closure. The findings of the literature review identify access to mobile technologies, pedagogical factors like effective usage of technology for teaching and learning, and connectivity as the main barriers for schools in rural areas. Findings from this exercise are used as factors to be considered when building a resilience framework for introducing and using mobile technologies in South African rural schools. The purpose of undertaking such a task is to complete a continuous evaluation of the need to develop a resilience framework and guidelines for the introduction and use of mobile technologies in South African rural schools.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116612458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845519
J. Osakwe, G. Iyawa, M. Ujakpa, Jovita Mateus
The aim of this paper was to present the findings from an empirically grounded investigation into the perception of Students on gamifying learning in high schools within African countries, informing the current literature on gamification in learning within African contexts and providing a new approach on how gamification can be incorporated into learning within African countries namely Namibia, Nigeria, Ghana and Kenya. The study was conducted in four countries in Africa. 800 quantitative survey questions were distributed among high school students in these countries. However, 525 questionnaires were completed. Simple random sampling technique was used. Descriptive Statistics was used to analyse the findings using the Statistical Package for Social Sciences (SPSS). The findings revealed that students are already actively involved in gamified learning activities in class and found them useful for their learning. However, there is a need to look at the distractive tendencies it poses in class. Recommendations on how to incorporate gamification in learning within African countries are proposed.
{"title":"Gamifying Learning in High Schools: Perceptions of Students in Selected African Countries","authors":"J. Osakwe, G. Iyawa, M. Ujakpa, Jovita Mateus","doi":"10.23919/IST-Africa56635.2022.9845519","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845519","url":null,"abstract":"The aim of this paper was to present the findings from an empirically grounded investigation into the perception of Students on gamifying learning in high schools within African countries, informing the current literature on gamification in learning within African contexts and providing a new approach on how gamification can be incorporated into learning within African countries namely Namibia, Nigeria, Ghana and Kenya. The study was conducted in four countries in Africa. 800 quantitative survey questions were distributed among high school students in these countries. However, 525 questionnaires were completed. Simple random sampling technique was used. Descriptive Statistics was used to analyse the findings using the Statistical Package for Social Sciences (SPSS). The findings revealed that students are already actively involved in gamified learning activities in class and found them useful for their learning. However, there is a need to look at the distractive tendencies it poses in class. Recommendations on how to incorporate gamification in learning within African countries are proposed.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122391368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845581
Sarathiel Chaipa, E. Ngassam, Singh Shawren
This paper discusses the outcome of combining insider threat agent taxonomies with the aim of enhancing insider threat detection. The objectives sought to explore taxonomy combinations and investigate threat sophistication from the taxonomy combinations. Investigations revealed the plausibility of combining the various taxonomy categories to derive a new taxonomy. An observation on category combinations yielded the introduction of the concept of a threat path. The proposed taxonomy tree consisted of more than a million threat-paths obtained using a formula from combinatorics analysis. The taxonomy category combinations thus increase the insider threat landscape and hence the gap between insider threat agent sophistication and countermeasures. On the defensive side, knowledge of insider threat agent taxonomy category combinations has the potential to enhance defensive countermeasure tactics, techniques and procedures, thus increasing the chances of insider threat detection.
{"title":"Towards a New Taxonomy of Insider Threats","authors":"Sarathiel Chaipa, E. Ngassam, Singh Shawren","doi":"10.23919/IST-Africa56635.2022.9845581","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845581","url":null,"abstract":"This paper discusses the outcome of combining insider threat agent taxonomies with the aim of enhancing insider threat detection. The objectives sought to explore taxonomy combinations and investigate threat sophistication from the taxonomy combinations. Investigations revealed the plausibility of combining the various taxonomy categories to derive a new taxonomy. An observation on category combinations yielded the introduction of the concept of a threat path. The proposed taxonomy tree consisted of more than a million threat-paths obtained using a formula from combinatorics analysis. The taxonomy category combinations thus increase the insider threat landscape and hence the gap between insider threat agent sophistication and countermeasures. On the defensive side, knowledge of insider threat agent taxonomy category combinations has the potential to enhance defensive countermeasure tactics, techniques and procedures, thus increasing the chances of insider threat detection.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127709323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-16DOI: 10.23919/IST-Africa56635.2022.9845637
Avashna Govender, D. Paul
Voice conversion (VC) is an important technique for the development of text-to-speech voices in the use case of lacking speech resources. VC can convert an audio signal from a source speaker to a specific target speaker whilst maintaining the linguistic information. The benefit of VC is that you only require a small amount of target data which therefore makes it possible to build high quality text-to-speech voices using only a limited amount of speech data. In this work, we implement VC using a mel-spectrogram Generative Adversarial Network called MelGAN-VC. This technique does not require parallel data and has been proven successful on as little as 1 hour of target speech data. The aim of this work was to build child voices by modifying the original one-to-one MelGAN-VC model to a many-to-many model and determine if there is any gain in using such a model. We found that using a many-to-many model performs better than the baseline one-to-one model in terms of speaker similarity and the naturalness of the output speech when using only 24 minutes of speech data.
{"title":"Multi-MelGAN Voice Conversion for the Creation of Under-Resourced Child Speech Synthesis","authors":"Avashna Govender, D. Paul","doi":"10.23919/IST-Africa56635.2022.9845637","DOIUrl":"https://doi.org/10.23919/IST-Africa56635.2022.9845637","url":null,"abstract":"Voice conversion (VC) is an important technique for the development of text-to-speech voices in the use case of lacking speech resources. VC can convert an audio signal from a source speaker to a specific target speaker whilst maintaining the linguistic information. The benefit of VC is that you only require a small amount of target data which therefore makes it possible to build high quality text-to-speech voices using only a limited amount of speech data. In this work, we implement VC using a mel-spectrogram Generative Adversarial Network called MelGAN-VC. This technique does not require parallel data and has been proven successful on as little as 1 hour of target speech data. The aim of this work was to build child voices by modifying the original one-to-one MelGAN-VC model to a many-to-many model and determine if there is any gain in using such a model. We found that using a many-to-many model performs better than the baseline one-to-one model in terms of speaker similarity and the naturalness of the output speech when using only 24 minutes of speech data.","PeriodicalId":142887,"journal":{"name":"2022 IST-Africa Conference (IST-Africa)","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132670880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：