Next Article in Journal
Hamas’s Hostage Videos as a Tool of Strategic Communication
Next Article in Special Issue
Localising the Creator Economy: How South African Student Influencers Adapt Global Monetisation Strategies on TikTok and Instagram
Previous Article in Journal
Seeing Bias at a Glance: A Visual–Statistical Analysis of Sentiment in China’s State-Backed English News Media
Previous Article in Special Issue
Humor That Hurts: An Exploration of Jokes About Black Women with Disabilities on TikTok in South Africa
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Exploring the Nexus of Opportunities and Challenges in Indigenous Language Podcasting Through Natural Language Processing of User-Generated Content

by
Bukola Christiana Ajala
1,2,*,
Abiodun Salawu
1,
Israel Ayinla Fadipe
1,3 and
Yetunde Pesu Aromavo
4
1
Indigenous Language Media Research Entity, North-West University, Mahikeng 2745, South Africa
2
Department of Mass Communication, Faculty of Arts, Social and Management Sciences, Dominion University, Ibadan 200243, Nigeria
3
Department of Mass Communication, and Media Studies, Faculty of Arts, Humanities, Management and Social Sciences, Augustine University, Lagos 106103, Nigeria
4
Department of Computer Science, Faculty of Computing and Applied Science, Dominion University, Ibadan 200243, Nigeria
*
Author to whom correspondence should be addressed.
Journal. Media 2025, 6(4), 179; https://doi.org/10.3390/journalmedia6040179
Submission received: 31 July 2025 / Revised: 5 October 2025 / Accepted: 8 October 2025 / Published: 16 October 2025

Abstract

Part of the relics of colonialism on the African continent is the loss of social identity caused by the adoption of colonial languages, leading to the endangered status of indigenous African languages. This qualitative study examines the potential and challenges of podcasting in indigenous African languages, with a focus on Yoruba. We conducted a sentiment analysis of the podcast “I Speak Yoruba Too” and “learn Yoruba online” to assess the range of audience feedback on the podcast. 735 data points were gathered and preprocessed, Hugging face transformers were used to analyse the sentiments on audience feedback. The result of the analysis shows that the negative reviews were 183, the neutral reviews 226, and the positive reviews 326. The visualisation of the word cloud of the labels shows the words frequently used in the reviews, revealing the challenges and the appreciation of the commenter. An in-depth interview was conducted with the host of the “I Speak Yoruba Too” podcast and the “learn Yoruba online Podcast”. The findings reveal that part of the challenges of podcasting include the absence of a standard Yoruba curriculum for foreign learners and time constraints. This paper argues that the deterministic nature of podcast technology offers opportunities to content creators and listeners, based on the medium’s flexibility and ease of access in facilitating language acquisition. Audience reviews and interview results also confirm the potential of the podcast to generate community building and social identity formation among learners. However, the monetisation of such digital products is often underexplored by both emerging and established podcasters.

1. Introduction

A Podcast is a marriage of two words: iPod and broadcast. The term podcasting was coined by Hammersley in 2004 in a newspaper article titled Audible Revolution (Yaman, 2016; Sharon, 2023). It is an audio file usually in MP3 format, downloadable from the internet to a computer or other media player; a series of such files can also be received through subscription (Sharon, 2023). Podcasting was popularised with the advent of online streaming services and the development of Apple and the upgrade of iTunes to 4.9. The ease of production, low-cost production, flexibility, and mobile streaming are part of why podcasts are in high demand by content creators and consumers (Panagiotidis, 2021; Andok, 2025; Dhiman, 2023). Podcasts were initially restricted to audio broadcasts but now encompass audio and video productions. Studies in the literature reveal that podcasts offer tremendous benefits for students’ learning. It has been used with remarkable success, particularly in distance learning programmes (Makina, 2020b; Forbes & Khoo, 2015; Newman et al., 2021; Makina, 2020a). Without a doubt, emerging technologies play a huge role in the preservation and promotion of indigenous language adoption.
A leading advocate for African indigenous languages, Abiodun Salawu argues that the pedagogical role of English in educational training in former British colonies has contributed to the decline in the use of African indigenous languages (Salawu, 2006). For instance, Fabuni and Salawu (2005) argue that the educated elite promote the adoption of the English language among their children and consider many hitherto Yoruba forms of greeting a complete waste of time. Meanwhile, a people’s language reflects their identity (Onadipe-Shalom, 2024), which leads to the need to revive the declining state of the Yoruba language to avert the danger of language extinction. In a study on language contact and language conflict among Yoruba-English bilinguals in Nigeria, Dada (2006) opines that speakers of all ages almost exclusively use the English language for official purposes. For unofficial matters, the older generation used the Yoruba language more, while the youth used either both languages or English more. The study concluded that youth are not proficient in the use of the Yoruba language, which portends dire consequences for the survival of the language. Code switching and code mixing between English and Yoruba have led to the disappearance of certain Yoruba lexicon.
Meanwhile, existing scholarship on emerging technologies and language revitalisation reveals a preponderance of studies on social networking platforms and the preservation of indigenous languages (Mapulane et al., 2024; Edyangu, 2021). We also observe that scholars seem to be caught up in the AI buzz, as there are suggestions for incorporating Augmented and Virtual Reality in indigenous language revitalisation. This leads us to the concept of Natural Language Processing (NLP), which is a branch of artificial intelligence that enables computer systems to decipher and comprehend human languages such as Yoruba, English, Chinese, Spanish, and others. NLP encompasses technologies such as speech recognition, information extraction, automatic text generation, speech synthesis, machine translation, and sentiment analysis (Dong, 2025).
The abundance of big data on social media platforms and the need to analyse such data for marketing and related decisions have made it essential for companies to adopt Natural Language Processing for sentiment analysis of customer reviews. Sentiment analysis applies to different aspects of human endeavour; it is concerned with understanding netizens’ online evaluations of products and services. Described as the extraction of subjective information through data mining and natural language processing, it helps draw useful conclusions about the polarity of such judgments (Mejova, 2009). It offers a cost-efficient method for analysing a huge data set online (Mao et al., 2024). Sentiment analysis has been beneficial in managing customer relationships, modifying/developing products, and facilitating customer satisfaction.
Regarding the sentiment analysis of the content of podcasts and audience reviews of podcasts, considerable research has been conducted on the analysis of health podcasts (Dumbach et al., 2024; Tabassum et al., 2019), political podcasts (Rizwan et al., 2025) and podcasts teaching the English language. However, there is little research on the use of sentiment analysis of indigenous language podcasts. Meanwhile, unlike other products and services, podcasts are rarely advertised. Typically, people become aware of podcasts through word-of-mouth communication and/or audience reviews on social media platforms. A potential consumer of an online product tends to trust the views of prior consumers more than available marketing information (Neri et al., 2012). Customer trends and customer feedback can be measured through comments and reviews on social media. Hence, the importance of sentiment analysis of audience reviews and comments on Yoruba podcasts. Sentiment analysis of African indigenous language podcasts is essential as it will provide information that will improve product quality, define marketing strategies, and improve customer relations (Selimovic et al., 2024). Extracting data on user opinions on social media can be a daunting task; therefore, sentiment analysis offers the opportunity for easy and rapid analysis of users’ subjective opinions expressed on social media platforms related to Podcasts. It can also provide information on understanding users’ motivations to engage with African indigenous language podcasts such as Yoruba language Podcasts. Motivations for listenership to podcasts according to the literature are companionship, interactive listening, knowledge acquisition, entertainment, convenience, and multi-tasking (Li et al., 2024).
Considering the potential of sentiment analysis in the analysis of big data opinions on social media platforms, we seek to analyse the users’ sentiments of selected Yoruba language podcasts that teach language acquisition. Yoruba is a language of the Niger-Congo family and is spoken predominantly among indigenous peoples of Southwestern Nigeria. Its speakers extend from the country to Togo, Benin Republic, Gambia, Ghana, Sudan, Sierra Leone, Côte d’Ivoire, and some parts of Cuba, Brazil, Trinidad and Tobago (Fabuni & Salawu, 2005; Fasinro et al., 2024). The Yoruba-speaking people heavily populate a village named Oyotunji (Oyo Renaissance) in South Carolina, Cuba, and Saint Dominque (Ajala, 2025). The language has many exonyms (outside names), such as Yorubawa, Nago, Yariba, Anago, Lukunmi, and Aku, possibly because of the diverse societies where it has spread. The great transatlantic slave trade around the mid-16th century, for example, led to the migration of Yoruba-speaking people into Brazil (Washington, 2016; Uguru & Okeke, 2020).
The relics of colonialism, coupled with the advent of globalisation and poor cultural pride among the youth, have led to the Yoruba language’s attrition rate (Salawu, 2006). The spread of the Yoruba language is deceptive, as according to UNESCO’s categorisation of endangered languages, Yoruba falls within the category of a ‘definitely endangered’ language, given that children no longer learn the language as a mother tongue in the home (UNESCO, 2025), Nonetheless, emerging technologies such as podcasting have been identified as critical for language learning, preservation and revitalization (Pandey, 2024). This perhaps explains why institutions of higher learning such as the University of Wisconsin-Madison in the United States recommend to learners’ podcasts such as I speak Yoruba, Too. This is of particular interest to these researchers; hence, we seek to identify the opportunities and challenges in Yoruba podcasting not only from the perspective of listeners/viewers as is common in the literature but also from the perspective of the host.
While previous studies have identified the motivations for audience use of podcasts (Z. Lin & Ng, 2025; Tobin & Guadagno, 2022), there is scant literature on the sentiment analysis of low-resource African indigenous languages and how African language podcasts shape audience social identity. We, therefore, examine how two purposively selected Yoruba podcasts shape the audience’s social identity. The I speak Yoruba too podcast was selected as a case study, and the learn Yoruba online podcast was also sampled alongside to triangulate the findings in the study. The findings are expected to generate robust data on audience feedback on indigenous language podcasting. This study is unique because of its multi-stakeholder perspective on indigenous language podcasting, as both the sentiment analysis of audience opinion on social media platforms and the results from content creator interviews will create a more nuanced understanding of the opportunities and challenges inherent in podcasts that teach indigenous African languages.

2. Research Objectives

The overarching objective of this study is to understand the feedback of listeners on the I Speak Yoruba Too podcasts through a sentiment analysis of users’ comments on social media platforms. More specifically, this study seeks to fulfil these objectives:
  • Identify the opportunities and challenges in Yoruba podcasting
  • Explore the various levels of sentiments expressed in the listener’s feedback on selected Yoruba language podcasts
  • Examine the specific ways Yoruba podcasts shape listeners’ social identity

3. Materials and Methods

A qualitative approach to the research was adopted, combining interviews with sentiment analysis of the I Speak Yoruba Too and Learn Yoruba Online Podcast.

3.1. Data Collection for In-Depth Interview of Content Creators

The first phase of the study involved an in-depth interview with the creator of the I Speak Yoruba Too podcast, as well as the Learn Yoruba online Podcast. A semi-structured interview guide guided the conduct of the interview. The rationale for the I Speak Yoruba Too Podcast is premised on its focus, which is to ensure that learners acquire the language seamlessly, and its acceptance as a learning aid at the University of Wisconsin in the United States. Hence, the I Speak Yoruba Too podcast was purposively sampled in the study as a case study, while the Learn Yoruba online podcast was sampled to triangulate the findings of the study. The creators of the content were informed of the study’s objectives before the interview, and an informed consent form was signed. Participation was voluntary, and the interviewee received no financial incentives.
The first episode of the I Speak Yoruba Too podcast was produced in 2022, although the Yoruba lessons were initially streamed on YouTube shortly after the pandemic. The podcast is hosted and produced by Eniola Ogunleye, a native speaker of the language, whose inspiration was initially sparked by boredom and a desire to promote the uptake of the Yoruba language among both native and non-native speakers. The podcast has a presence on various social media platforms, including YouTube, Instagram, and Facebook. It is published on Spotify, Apple Podcasts and on Red Circle (the domain page of I Speak Yoruba Too. The second podcast that was purposively sampled for this study is the learn Yoruba online podcast, which debuted in 2021. The host of the podcast started the initiative based on two premises, namely: the belief that he had the expertise to analyse the language more efficiently than available content creators on social media and the passion to impart knowledge of the language to others.
Ten semi-structured interview guides that allowed follow-up questions guided the conduct of the interview. The questions were formulated to satisfy the objectives of the study and pertained to eliciting responses on the opportunities of podcasting to content creators and audiences, the challenges faced by individual podcast hosts, how the challenges are mitigated, how podcasting engenders community building, the formation of social identities and parasocial relationships. The interviews were conducted through Google Meet and recorded on a mobile device. One of the researchers transcribed it, and the data was cleaned to facilitate readability. The results were thematically analysed through the process of familiarisation with the data, sifting through the data, identifying emerging themes, coding the themes, and checking the themes. The themes are presented in the findings section.
Table 1 shows the research objectives and methods adopted to achieve each objective.
The second phase of the study involved collecting audience reviews of the two sampled podcasts on selected social media platforms. The data were analysed on hugging face transformers. The analysis was used to understand the opinions of the users, challenges, appreciation, and how the podcast can be improved for efficient learning.

3.2. Data Collection for the Sentiment Analysis

We gathered 735 data sets on the I Speak Yoruba Too podcast and learn Yoruba online. Reviews were collected from social media platforms, YouTube, and Spotify. Both social media platforms were selected primarily because existing and potential listeners of the podcast typically access those sites to download, listen to, or watch the podcast. The data gathered from the I speak Yoruba Too podcast were 307 and Learn Yoruba Online has a total of 428. The two data sets were merged to make a total of 735 comments.

3.3. Preprocessing of Data

The data shape was (735, 4), 735 rows and 4 columns. The 4 columns were 1. Name (i.e., the name of the commenter). 2. Published, which is the date the comment or review was published. 3. Comment, which is the content of the review and 4 Like, which is the number of likes the comment received. All other columns were deleted, except for “Comment”. The shape of the new data set was (735, 1), having 735 rows and a column for comments. Some typographical and grammatical errors were corrected in the comments to facilitate readability.

3.4. Data Analysis

A hugging face transformer called Cardiffnlp-twitter-roberta-base-sentiment was used to analyse the data. The Cardiffnlp-twitter-roberta-base-sentiment is a pertained model based on the Roberta model, and it was trained on 58 million tweets and specifically refined for sentiment analysis (Barbieri et al., 2020). The Cardiffnlp-twitter-roberta-base-sentiment was used because it is a three-class sentiment classifier having negative, neutral, and positive labels. A sentiment pipeline was created, and a dictionary of labels and scores was created to store the predicted number of labels in the data with their scores. The sentiment analysis can be found at https://www.kaggle.com/code/yetundepesu/yoruba-sentiment-analysis-1 (accessed on 7 October 2025).

4. Results

The results are presented according to how they address the research objectives. The first objective was to identify the opportunities and challenges in Yoruba podcasting. Some themes (patterns of responses) that emerged from the analysis of the interview and sentiment analysis are (1) opportunities in Yoruba podcasting for content creators with such sub-themes (a) Ease of production, (b) Financial incentives, (2) Opportunities in podcasting for listeners with the sub-theme: facilitate language acquisition and ease of access to podcast. The other part of the first research objective deals with the challenges in Yoruba podcasting, and the pattern of response from the interview and the sentiment analysis were presented along the sub-themes (a) lack of a curriculum for native speakers of the language and (b) contentions regarding the standard usage of the language.
For the second research objective, which deals with identifying the various levels of sentiments in audience reviews of sampled podcasts. The results are presented in tables. The third research objective sought to examine the specific ways Yoruba podcasts shape users’ social identity, and the only sub-theme here is community building. Each of these is discussed in turn, starting with the first.

4.1. Opportunities in Yoruba Podcasting for Content Creators

4.1.1. Ease of Production and Free Accessibility

We observe from the responses of the content creators that podcasting has potential for diverse categories of users. This potential perhaps explains its worldwide acceptance and use in recent times. For content creators, technology facilitates easy production and offers free accessibility. In the words of one of the interviewees:
Podcasting makes it easier to access content creation; if you are following the social media world, anyone can produce a podcast. All you need to do is set up your phone and microphone and then you have a podcast. For content creators, it would be amazing to see more people producing podcasts because of their free and easy accessibility.
However, for the host of the learn Yoruba online, podcasting offers an opportunity for content creators to diversify their revenue base and perhaps showcase the tourism potential of their country. According to him,
Through podcasting, I tell listeners stories about Yoruba-speaking people. I also fly drones to show aerial pictures of places I have been to in Nigeria. The fact that I show interesting fun spots such as Erin Ijesa waterfalls sparks the interest of viewers in the diaspora and this motivates their interest in visiting the country. For example, there is a woman whom I teach in the United States who is planning a vacation to Nigeria for next year. I will act as her tour guide.

4.1.2. Financial Incentives

The influx of various individuals into the act of podcasting, from young talents eager to share their stories to professionals who perhaps create podcasts to diversify their income, indicates that it is a profitable venture. Podcasting provides a platform for advertising products or services. Content creators are rewarded for their long hours of production, while companies benefit from promoting products directly to specific market segments. According to one of the interviewees:
The financial incentive is something I am just getting into myself, because when all my Yoruba initiatives started, they were all free. From there, I researched the marketplace to find what is lacking. What do I need to do, what do I need to add to my content, and how can I monetise it? I have been able to find my niche and what I need to do, which is helpful. I believe that when you give your best, people will buy whatever you have to offer. Even when there is a price tag, they will appreciate that they are getting value for their money. With the I Speak Yoruba Too podcast, I have been able to get many students who give positive feedback on the quality of my podcast and its impact. For me, that is one of the incentives for a Yoruba teacher. I teach Yoruba on a podcast. It’s a good way for one to build a client; students come in from different places and give endorsements. From there, people would also seek collaboration.
The foregoing position on the financial opportunities in podcasting was reiterated by another interviewee, who argued that
The primary advantage podcasting offers me is in terms of financial reward. I have been able to acquire private students through podcasting. The private students I teach pay per hour. Last week, YouTube unlocked a feature in which creators can be rewarded. This feature is tagged ‘Super thanks’ on YouTube. So, one of my viewers gave me a ‘Super thanks’ of 10 dollars in appreciation for one of my uploaded videos on Youtube. People tend to appreciate podcasts and reward content creators.
The business potential of podcasts is not limited to dynamic ad insertion; search engines also increase the visibility of podcasts through recommendations (Aufderheide et al., 2020). Algorithms aid the discoverability of podcasts, as users’ chat history on online platforms determines available promotions on their feeds.

4.2. Opportunities in Yoruba Podcasting for Listeners

Facilitate Language Acquisition and Ease of Access to Podcasts
Podcasts are described as an extremely versatile technology in which content is tailored to meet the needs of users and can be accessed without restrictions on time and location (Andok, 2025). The potential in podcasting for learning the Yoruba language is expressed in the different sentiments presented in Table 2.
Table 2 describes the various subjective opinions expressed by users on the I Speak Yoruba too Podcast. The remarks labelled positive and neutral sentiments reveal much information about the value listeners derive from learning Yoruba through Podcasts. Some of the reviews are instructive for the host to better package her content to meet identified needs. For example, in one of the comments, the host was advised to write on a screen to facilitate audience visual memory. Some of the sentiments expressed by the audience are consistent with the perspective of the host of the podcast.
While speaking on the potential of podcasts in Yoruba language teaching, the host of I speak Yoruba Too observed that
It would help the Yoruba community, especially those interested in learning to speak Yoruba and who are oblivious of where to start. From my perspective, it is a great way for anyone who wants to learn Yoruba to start; they would benefit enormously from it. My podcast lessons are structured so that beginners learn the rudimentary aspects of the language and then gradually build on them. I have received a few emails informing me of how much my podcast has helped them. This calls for deep introspection on the part of those considering creating content through podcasts. One must not venture into podcasting purely for the monetary incentive but based on value creation.
While the other interviewee posited that indigenous language podcasting indeed offers listeners/viewers the opportunity of language acquisition, he stressed the added value of social identity formation. According to him,
Recently I was speaking with a student in the United States who expressed deep regret that, even though she has a Yoruba name, she does not understand the language. For her, that led to an identity crisis as her name was not a true representation of her identity. She is of black descent, and her mother, in a quest for social identification, decided to give her child a Yoruba name based on her knowledge of the culture and her past relationship with a Yoruba man. The child was therefore faced with the challenge of explaining to anyone she came across that she did not understand the language although she has a Yoruba name. Although she lives in the US, she has an identity crisis as a black US citizen who cannot trace her roots. For this category of people, learning Yoruba is a big deal, because it can at least constitute a sense of identity for them. Similarly, some people in the diaspora who are born to Nigerian parents but cannot speak the language place a premium on learning the language as a means of social identity.

4.3. Challenges of Yoruba Podcasting

4.3.1. Lack of a Curriculum for Non-Native Speakers of the Language

Typically, when podcasts are used as learning aids in educational settings, teaching is tailored to suit the requirements of a curriculum. This perhaps explains why the absence of a standard Yoruba curriculum for second language learners of the language was described as a major challenge confronting Yoruba language teachers. According to one of the interviewees:
The first significant challenge is the absence of a curriculum for foreign learners of the Yoruba language. Concerning learning English as a second language, for example, there is a format for it in respect of a curriculum for beginners. Unfortunately, that is not the case when learning Yoruba as a foreigner. There is a Yoruba curriculum for Yoruba speakers in Yoruba land. However, there is no curriculum for second language learners who have had no prior exposure to the language. Hence, my Yoruba podcast is built on a Korean curriculum. My lesson is based on my experience as a beginner while learning the Korean language. I used Korean tools for beginners as a template to research teaching the Yoruba language to beginners. In that way, I developed my curriculum. I observed that the starting point for beginners learning a new language is how to greet people. So, I replicated this in my curriculum for the Yoruba podcast. It will be good to have a standard format for teaching second language learners how to speak Yoruba. For example, the way you learn English in England is the same way it is learnt in other countries, such as the United States or Canada. Although there may be differences in accent or nuances, they are all the same.
However, the perceived challenge of the other interviewee is not related to curriculum development. He stated that,
The specific challenge I experienced has to do with time constraints. Initially, I had to contend with juggling my responsibilities in a paid job with creating podcasts, so I was faced with time constraints at the time. Although I am no longer in paid employment, now I am a full-time content creator. However, the tonal nature of the Yoruba language makes the editing of the board time consuming. It takes a lot of time to edit. I was also doing 9-5 work to increase my finances at the time.

4.3.2. Comments on the Standard Yoruba Use

Closely associated with the issue of curriculum development is the challenge regarding the standard usage of the language. For one of the interviewees,
Concerning the Yoruba language, there are contentions regarding the standard usage of the language. I observed in the little research I conducted on the Yoruba language that the standard Yoruba language was last revised in the 1990s, which is a long time ago. One of the areas where there are disagreements is the use of the apostrophe in the Yoruba language. Although some people support its use, others do not. Yoruba is a tonal language and for me, the apostrophe does not perform a function in the language. That is one of my challenges; despite this, I research widely in language to ensure that I don’t mislead my audience.
However, the other interviewee did not express a similar sentiment. He noted that the major challenges he has faced as a content creator are time constraints and finance. In his words,
“It takes a lot of time to edit videos due to tonal marks. I was also in paid employment to supplement my finances at the time”.
This foregoing position on the creation of the Yoruba podcast aligns with some of the feedback received from the audience on the quality of the content. From the sample in Table 3, it is clear that aside from the tonal marks that may make video editing challenging, content creators may not pay attention to other indices that improve the quality of production, such as acoustics, vocal pitch, and the need to adapt content to a wide variety of audiences. Table 3 presents information on some of the challenges faced by the audience while listening to the Yoruba language podcast. The challenges range from poor presentation of content to low-quality audio content.

4.4. The Various Levels of Sentiments in Audience Reviews

The results of the sentiment analysis show that of the 735 reviews, 356 were positive, 224 were neutral, and 155 were negative, as shown in Table 4. The negative score of 76.75% shows that 155 reviews have 76% likelihood to be negative reviews, the neutral score of 72.39% shows that 224 reviews have 72% likelihood to be neutral reviews and the positive score of 89.93% shows that 356 reviews have 89% likelihood to be positive reviews.
The WordCloud library was used to quantify the frequency of words in the labels Negative, Neutral, and Positive. WordCloud shows word clusters for each review class, highlighting the usage of words in portraying the challenges, appreciation, and encouragement of the podcast.

4.4.1. Negative Sentiments

The negative score of 76.75% shows that 155 reviews have a 76% probability of being negative reviews. The analysis in Table 4 shows that the negative sentiment is the lowest, denoting the dissatisfaction of the reviewers with the podcast. The word cloud in Figure 1 shows words like Yoruba, podcast, quality, enjoy, hard, difficult, sound, etc., which were the words used by the reviewers to complain about the technical issues, difficulty of the language, clarity and detail of the podcast.

4.4.2. Neutral Sentiments

The neutral score of 72.39% shows that 224 reviews have a 72% likelihood of being neutral reviews. The neutral sentiments from Table 4 are not necessarily negative or an expression of reviewers’ dissatisfaction but are reviews that are certainly not impactful. The review may be requesting clarification or more lessons. The word cloud in Figure 2 shows the word cluster used in the neutral sentiment, examples are Yoruba, please, word, language, and Video.

4.4.3. Positive Sentiments

The positive score of 89.93% shows that 356 reviews have 89% likelihood to be positive reviews. The analysis, in Table 4, shows that the positive sentiments are the highest, denoting satisfaction and appreciation from the reviews. From the word clouds in Figure 3, words like Thank, Yoruba, helpful, Best, good, great, video were used to convey that more learners appreciated and understood the podcast.

4.4.4. The Various Levels of Sentiment Expressed in the Listener’s Feedback on “I Speak Yoruba Too

The field of sentiment analysis is not only critical in understanding human-computer interaction, but it also provides insights into public opinion and trends. Nonetheless, it is challenged by complexities such as the variability in human emotions accentuated by language diversity and dynamism(Alahmadi et al., 2025). In this study, we present the various levels of audience sentiments on the I Speak Yoruba Too and the learn Yoruba online podcast. The podcast reviews are presented in Section 4.4.5

4.4.5. Podcast Reviews

The identified subjective evaluation of the reviews is along positive, neutral and negative dimensions, and these are presented in Table 5a,b.
Table 5a,b are samples of audience value judgements expressed in positive, neutral, and negative sentiments. We observe from the pattern of the positive reviews that the areas of interest expressed in the podcast are related to the quality of the teaching, the content of the podcast, and the vocabulary acquisition of learners. Specific references were made concerning the informative nature of each podcast that serve not only as a source of marketing to prospective listeners but will certainly inspire a more excellent delivery from content creators. The nature of audience feedback as it relates to both podcasts is largely the same. Apparently, both podcasts attract not only native but non-native speakers of the language spread across different geographical boundaries. This has potential for cross-cultural communication.
However, some of the comments categorised as positive reviews lack information on the specific aspect of the podcast which the audience finds most appealing. Such phrases as: “this is amazing! Thanks a lot, we appreciate you, you are too much” provide little content information on what is outstanding in the podcast. For the learn Yoruba Online Podcast, it is interesting that one of the negative reviews had to do with the facial disposition of the content creator. Another negative feedback for that podcast showed that the listener/viewer did not even learn how Yoruba is spelt. Meanwhile, learners gave useful criticism concerning the I Speak Yoruba Too Podcast, particularly in instances where the host was described as repetitive and the content/quality of delivery was decried. Negative reviews are, as a matter of fact, extremely important as those sentiments may perhaps motivate the host to go back to the drawing board and devise better strategies to satisfy the audience.

4.5. The Specific Way Yoruba Podcasts Shape Listeners’ Social Identity

Community Building

The path for the formation of social identities and the building of communities between creators and users was also explored in the interviews. According to one of the interviewees, her podcasts engender a sense of parasocial relationship among listeners. In her words:
The podcast has encouraged diverse interests, fostering community building among people. My students typically ask questions like “How can we find a language exchange? Who can we practice speaking Yoruba with?” Meanwhile, I have friends who also teach Yoruba. They organise Yoruba-related events to increase learners’ engagement with the language. This helps learners practise speaking Yoruba. Students often struggle to speak the language. For example, when they mispronounce a certain word, they feel discouraged and tend to withdraw. Therefore, we provide a community that acts as a safe space to encourage them to pronounce words in any way that they can. These communities exist in two forms: physical and online. I am not directly involved with the physical community. I have friends who help me organise physical events whenever such occasions arise.
However, the other content creator interviewed expressed a different perspective on the issue of how podcasts shape audience social identity formation
We have a mass exodus of youths travelling to the United Kingdom for further studies; thus, migration has allowed youths of different background to find a common ground in the diaspora. Unfortunately, some of the youths do not understand the Yoruba language despite their background. Thus, when such youths meet their peers from the same ethnic affiliation, they will seek to bond with them through the instrument of language. Many people are bringing the Yoruba language into the community abroad. As this is done, some people will feel secluded and the desire for inclusion will spur them to seek to acquire the language which is a unit of identification by ingroup members. Before the influx of these migrants, many youths did not see the importance of learning the language because they did not meet many Nigerians who understood the language. But now, the fact that there are more Nigerian migrants in the United Kingdom has made it imperative that other Nigerians who felt ostracised from the language seek to learn it to feel included.
The pattern of audience feedback on social media platforms regarding the two podcasts also indicates that a community is fostered amongst learners through the podcast. For example, from the pattern of feedback, we observe that the audience motivates engaging discussions in the comment section by posing questions such as: “Hello everyone, hope you enjoy this lesson. How many Yoruba phrases do you know? Share in the comment section”. Such remarks will foster a sense of community building among learners and will generate the formation of social identities.

5. Discussion of Findings

Although there is explosive growth in podcast technology globally; however, African uptake and consumption of media have been slower compared to countries in the Global North (Royston, 2023). From 2014 to 2018, when podcasting gained momentum in other parts of the world, in many parts of Africa, the platform was not embraced due to limited broadband for connectivity, a preponderance of feature phones and the availability of FM frequencies on such phones. Initially, smartphone internet subscriptions were limited to 3G and 4G networks which did not support online streaming or massive downloads of podcast content. This development affected the adoption of podcasting on the African continent. The limited broadband connectivity in African countries also contributed enormously to the digital divide between rural and urban areas leading to the inaccessibility of digital products in remote areas. Users preferred to listen to the radio on their phones rather than downloading audio media files (Royston, 2023). With the increase in smartphone subscription rates, lower internet costs, the availability of 5G broadband and a burgeoning class of content creators, the podcast platform in Africa has experienced a remarkable shift.
The flexibility of the medium and its capacity to create intimacy with listeners explain its global spread and acceptance (Andok, 2025). However, there are concerns that the sustainability of podcasting as an industry for content creators in Africa may be challenged by advertising revenue or sponsorships, which may not be easy to secure (Mbai, 2025). The findings in this study show that podcasters experience financial-related challenges that may impact the quality of their productions. The results of the sentiment analysis reveal that part of the challenges audience experiences when listening/watching podcasts concerns its audio/visual quality and the absence of good quality delivery in some instances. These findings are instructive for content creators to improve the visibility of their podcast in online spaces.
The visibility of a podcast invariably impacts listenership and grows the online community for the podcast. This, according to our interviewee, expands one’s clientele. It may, however, be challenging for established and emerging Podcasters to monetise their products if they are amateurs in metadata optimization (Dhiman, 2023). Identifying information (metadata) such as titles, descriptions, and other information aids the discoverability of podcasts, especially when potential audiences use search engines. Podcast audiences typically have certain information needs when searching for a relevant podcast. Although the topical relevance of the podcast is usually of paramount importance, listeners may also consider such indices as the series of episodes, the importance of the host or guest in an episode, and the presentation style of the podcaster (Carterette et al., 2021).
In addition to that, podcast technology gained acceptance based on some of its benefits, which include, but are not limited to, its bite-sized learning advantages, its flexibility in time and location for content producers and listeners (Panagiotidis, 2021). The results of the interview and the sentiment analysis in this study are consistent with the position of previous scholars on the potential of podcasting in education, as it pertains to student knowledge acquisition (Balladares, 2020; Hernandez-Lopez & Mendoza-Jimenez, 2025). This is an opportunity that has yet to be fully tapped into by emerging content creators, as it has the potential to revive the uptake of indigenous African languages among youth who regard colonial languages as more prestigious.
Nevertheless, the issue of curriculum development in the Yoruba language may, however, be a disincentive to aspiring content creators who seek to revitalise the dwindling uptake of the Yoruba language. For Arohunmolase (2006), curriculum development is a lingering problem in Nigerian educational institutions for the instruction of second language learners. In a related study on teachers’ perceptions and the value of Yoruba as a medium of instruction in the primary three classes, the authors posit that the Yoruba language, like other African languages, is confronted with diverse challenges related to a lack of instructional materials and government support (Abidogun et al., 2015).
Aside from the inherent opportunities and challenges of podcasting in African languages, the result of this study is consistent with the literature on the potential of podcasts to engender a participatory culture and opportunities for social media engagement between content creators and users. This feedback mechanism fosters a sense of community, which is consistent with scholarly positions on the community-building role of podcasts (Mbai, 2025; García-Marín, 2020). Beyond the online advantage offered by podcasts, users may be drawn to a podcast based on perceived social identity similarity or distinctiveness (S.-W. Lin & Huang, 2024). Excerpts from audience reviews of the sampled podcasts confirm the proposition of uses and gratification theory as to how audience members typically select media products to satisfy psychological and social needs. Although our focus in this study is not to analyse audience motivations for listening to a podcast, such information will better assist content creators in appropriately targeting content. In this study, we observe that some of the comments suggest that users’ goal of social identity drives them to search for media products that satisfy this need. An example is given in this excerpt that reads:
I have been on my spiritual journey and realised that many of my roots are in Yoruba. And it’s surprising how fast and easy this language is coming to me. I found your channel on Spotify.
Another comment closely related to the former also expresses the user’s goal of mastering the Yoruba language, which must have motivated their subscription to the channel.
I subscribed to this channel today, ready to learn everything that has been taught here from day one, as I will be watching all the videos and taking notes as well. I look forward to becoming better at speaking Yoruba. Thank you for doing this.
Listerners’ need to identify with such perceived social identity ultimately translates into parasocial relationships and community building. The host of the I Speak Yoruba Too podcast admits to the existence of parasocial relationships between her and her audience when she notes
Therefore, we provide a community that acts as a safe space to encourage them to pronounce words in any way that they can. These communities exist in two forms: physical and online. I am not directly involved with the physical community. I have friends who assist me in organising physical events whenever such occasions arise.
However, for the other content creator, Nigerian immigrants in the Global North use the instrument of language to foster social identity amongst peers. In the words of the host of learn Yoruba online
Before the influx of these migrants, many youths did not see the importance of learning the language because they did not meet many Nigerians who understood the language. But now the fact that there are more Nigerian migrants in the United Kingdom has made it imperative for other Nigerians who felt ostracized from the language to seek to learn it to feel included.
Nonetheless, podcast technology also offers intimacy to listeners, which has helped build a strong sense of community among them (Dhiman, 2023). Similarly, the transmedia character of podcasts acts as a catalyst to create user engagement opportunities (García-Marín, 2020). In contrast, user participation in podcasts is limited by factors such as topicality of content, the asynchronous nature of podcasts, and the technological design of podcatchers (specific smartphone apps). The interface design of some podcatchers makes it impossible for users to contribute meaningfully to the discussions on serialised podcasts. The absence of a user interface that facilitates feedback on podcatchers compels listeners to use computer systems for feedback purposes (García-Marín, 2020). Concerning topics as a limiting factor, audience reviews of the podcast reveal that some topics were found to be difficult. This explains why negative sentiments such as Yoruuba is so difficult, I don’t understand, etc., were expressed in the audience feedback.
However, there are concerns that despite the rise in the use of sentiment analysis, low-resource languages such as Yoruba are difficult to analyse using natural language processing (Abegunde et al., 2019). In this study, however, the majority of the audience reviews on the podcast were written in English; hence, it was relatively easy to evaluate the text. One of the limitations of this study is that the web scraper used in harvesting the data online made it difficult to disaggregate audience reviews from host feedback. Although the host feedback did not negatively impact the study findings. Meanwhile, earlier studies have also argued that it is difficult to assess sarcasm and words that convey contradictory meanings when used in different contexts (Sharma & Goyal, 2023).
Today, the growth in spam reviews through Bots also poses a challenge to the analysis of a corpus of data using sentiment analysis. Nonetheless, sentiment analysis offers the advantage of processing big data in real-time to facilitate informed decisions. (Păvăloaia et al., 2019). Emotional reactions of consumers expressed on social media are said to influence potential purchase decisions, considering the exponential increase in people using mobile devices to exchange opinions online (Păvăloaia et al., 2019). There are also concerns that the I Speak Yoruba Too podcast has repetitive content; this would likely stifle listeners’ interest.

6. Conclusions

This study has fulfilled its main focus, examining the opportunities and challenges inherent in teaching the Yoruba language through a podcast. Although the deterministic nature of podcast technology offers enormous benefits for content creators and listeners, there are perceived challenges that may limit its effectiveness in teaching and learning African languages. Through a sentiment analysis of audience feedback and an interview with podcasters, we discovered that podcasting has immense potential, particularly in revenue generation, for the teeming number of Nigerian youths who are in search of elusive white-collar jobs. In addition, for listeners, the flexibility of the technology offers users the opportunity to learn Yoruba at their own pace. However, the challenges inherent in using podcasts to teach the Yoruba language are diverse. For one of the interviewed hosts, the lack of a standard curriculum for non-native speakers of the language and contentions regarding the standard usage of the language are a disincentive. In contrast, the other host alluded only to time constraints in packaging podcast content.
The potential of podcasts to shape listeners’ social identity was explored in this study, and the results of the interviews and sentiment analysis of audience reviews reveal that the online nature of interaction between podcasters and the audience enables learners to operate within a community that fosters inclusiveness in learning, engagement and formation of parasocial identities. This study underscores the critical role of audience use and gratification when they actively seek podcasts that teach language acquisition. Their goal may not be merely to acquire language but also to form a social identity. The practical implications of these findings are that coloniality may have led to the erosion of our language and identity as Africans. Nevertheless, Globalisation and the emergence of digital technologies hold promise as the orality of the podcast technology for immersive storytelling is reshaping the concept of pedagogy.

Author Contributions

Conceptualisation, B.C.A.; Methodology, B.C.A. and Y.P.A.; Software Y.P.A.: Validation, B.C.A., I.A.F. and Y.P.A.; Formal analysis, B.C.A. and Y.P.A.; investigation, B.C.A., A.S., I.A.F. and Y.P.A.; Resources, A.S., B.C.A. and Y.P.A.; Data curation, B.C.A. and Y.P.A., Writing-original draft preparation, B.C.A. and Y.P.A.; writing- review and editing I.A.F., B.C.A. and Y.P.A.; Visualisation, B.C.A., A.S. and Y.P.A.; Supervision, B.C.A. and A.S.; Project administration, A.S.; Funding, A.S., B.C.A., I.A.F. and Y.P.A. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Ethics Committee of the Dominion University Ibadan with protocol code DU/REG/ACAD/APPR/O2/VOL.1/25/06.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Abegunde, O., Iyanda, A., & Ninan, D. (2019). Design issues in sentiment analysis for Yorùbá written text. Ife Journal of Science and Technology, 3, 12–25. [Google Scholar]
  2. Abidogun, B. G., Phatudi, N. C., Joubert, I., & Hartell, C. G. (2015). Teachers’ perceptions and value of Yoruba as a medium of instruction in primary 3 classes. International Journal of Educational Sciences Special Edition, 8(1), 101–110. [Google Scholar]
  3. Ajala, B. C. (2025). Adejoke Ewaede Yoruba Archival System: Changing Media Narratives And Stereotypes of the Yoruba Woman. In A. K. Manfred, & T. D. A.-T. A. M. F. Asuman (Eds.), Communication and Social Change in Africa: Selected Case Studies (pp. 216–227). Institute of Network Cultures, Amsterdam University of Applied Sciences. [Google Scholar]
  4. Alahmadi, K., Alharbi, S., Chen, J., & Wang, X. (2025). Generalizing sentiment analysis: A review of progress, challenges, and emerging directions. Social Network Analysis and Mining, 15, 1–28. [Google Scholar] [CrossRef]
  5. Andok, M. (2025). Podcast—The remediation of radio: A Media theoretical framework for podcast research. Journalism and Media, 6, 7. [Google Scholar] [CrossRef]
  6. Arohunmolase, O. (2006). Globalization and the problems and prospects of teaching and learning of Yoruba as a second language (L2) in colleges of education in Nigeria. Available online: http://www.lingref.com/cpp/acal/36/paper1416.pdf (accessed on 7 October 2025).
  7. Aufderheide, P., Lieberman, D., Alkhallouf, A., & Ugboma, J. M. (2020). Podcasting as public media: The future of US news, public affairs, and educational podcasts. International Journal of Communication, 14, 22. [Google Scholar]
  8. Balladares, M. C. S. (2020). The perceived impact of podcasting in teaching scientific research: A study of student’s motivation and performance. Available online: https://www.researchgate.net/publication/380629137_THE_PERCEIVED_IMPACT_OF_PODCASTING_IN_TEACHING_SCIENTIFIC_RESEARCH_A_STUDY_OF_STUDENT’S_MOTIVATION_AND_PERFORMANCE (accessed on 7 October 2025).
  9. Barbieri, F., Camacho-Collados, J., Neves, L., & Espinosa-anke, L. (2020). TweetEval: Unified benchmark and comparative evaluation for tweet classification. In Findings of the Association for Computational Linguistics: EMNLP 2020 (pp. 1644–1650). Online. [Google Scholar] [CrossRef]
  10. Carterette, B., Jones, R., Jones, G. F., Eskevich, M., Reddy, S., Clifton, A., Yu, Y., Karlgren, J., & Soboroff, I. (2021, July 11–15). Podcast metadata and content: Episode relevance and attractiveness in ad hoc search. 44th International ACM SIGIR Conference on Research and Development in Information Canada Retrieval, Montreal, QC, Canada. [Google Scholar]
  11. Dada, S. (2006). Language contact and language conflict: The case of Yoruba-English bilinguals. In O. Ndimele, C. Ikekeonwu, & B. M. Mbah (Eds.), Language and economic reforms in Nigeria (pp. 68–85). M and J Grand Orbit Communications Ltd. and Emhai. [Google Scholar]
  12. Dhiman, B. (2023). Key challenges and opportunities for podcasters in the digital age: A critical review. MISC. [Google Scholar]
  13. Dong, Z. (2025). Exploring cross-cultural communication content adaptability through advanced natural language processing and sentiment analysis. Systems and Soft Computing, 7, 200290. [Google Scholar] [CrossRef]
  14. Dumbach, P., Schwinn, L., Löhr, T., Do, P. L., & Eskofier, B. M. (2024). Artificial intelligence trend analysis on healthcare podcasts using topic modeling and sentiment analysis: A data-driven approach. Evolutionary Intelligence, 17, 2145–2166. [Google Scholar] [CrossRef]
  15. Edyangu, H. (2021). Social media use and indigenous language preservation: A case study of Lukenyi language in Serere district–Uganda. College of Education Open, Distance and E-Learning of Kampala International. [Google Scholar]
  16. Fabuni, F. A., & Salawu, A. S. (2005). Is Yorùbá an endangered language? Nordic Journal of African Studies, 14, 18. [Google Scholar]
  17. Fasinro, K. S., Akinpelu, B., Alabi, O. B., Akinkuotu, F. A., & Onowugbeda, F. U. (2024). Evaluating the effectiveness of interactive radio instruction in teaching indigenous languages to native and non-native speakers in Nigeria. African Perspectives of Research in Teaching and Learning, 8, 92–104. [Google Scholar]
  18. Forbes, D., & Khoo, E. (2015). Voice over distance: A case of podcasting for learning in online teacher education. Distance Education, 36, 335–350. [Google Scholar] [CrossRef]
  19. García-Marín, D. (2020). Mapping the factors that determine engagement in podcasting: Design from the users and podcasters’ experience. Communication & Society, 33, 49–63. [Google Scholar] [CrossRef]
  20. Hernandez-Lopez, M., & Mendoza-Jimenez, J. (2025). Podcasts created by university students: A way to improve subject understanding, connection with peers, and academic performance. Education Sciences, 15, 284. [Google Scholar] [CrossRef]
  21. Li, Y., Wan, A., Hashim, H., Zhang, L., & Wu, A. (2024). Motivations for using podcasts: A systematic literature review. Information, Medium and Society, 22, 41. [Google Scholar] [CrossRef]
  22. Lin, S.-W., & Huang, C.-D. (2024). Hooked on audio! Unveiling the secrets of podcast stickiness through social identity and uses and gratification theories. Technology in Society, 76, 102422. [Google Scholar] [CrossRef]
  23. Lin, Z., & Ng, Y.-L. (2025). Unraveling gratifications, concerns, and acceptance of generative artificial intelligence. International Journal of Human–Computer Interaction, 41, 10725–10742. [Google Scholar] [CrossRef]
  24. Makina, A. (2020a). Developing a framework for managing the quality use of podcasts in open distance and e-learning environments. Open Praxis, 12, 67–81. [Google Scholar] [CrossRef]
  25. Makina, A. (2020b). Investigating the use of podcasts in an open, distance and e-learning environment. Perspectives in Education, 38(1), 30–42. [Google Scholar] [CrossRef]
  26. Mao, Y., Liu, Q., & Zhang, Y. (2024). Sentiment analysis methods, applications, and challenges: A systematic literature review. Journal of King Saud University-Computer and Information Sciences, 36, 102048. [Google Scholar] [CrossRef]
  27. Mapulane, M. G., Mangaka, A. C., Malatji, E. J., Baloyi, N. C., & Muthambi, R. (2024). Indigenous language preservation, challenges, and opportunities in the social media age. In Digital media and the preservation of indigenous languages in Africa: Toward a digitalized and sustainable society (p. 223). Lexington Books. [Google Scholar]
  28. Mbai, A. M. (2025). Podcasts as the new radio: Shaping the future of storytelling. African Multidisciplinary Journal of Research, 1, 551–563. [Google Scholar] [CrossRef]
  29. Mejova, Y. (2009). Sentiment analysis: An overview. University of Iowa, Computer Science Department, 5, 1–34. [Google Scholar]
  30. Neri, F., Aliprandi, C., Capeci, F., & Cuadros, M. (2012, August 26–29). Sentiment analysis on social media. 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, IEEE (pp. 919–926), Istanbul, Turkey. [Google Scholar]
  31. Newman, J., Liew, A., Bowles, J., Soady, K., & Inglis, S. (2021). Podcasts for the delivery of medical education and remote learning. Journal of Medical Internet Research, 23, e29168. [Google Scholar] [CrossRef]
  32. Onadipe-Shalom, T. (2024). The social media and the revitalisation of Yorùbá language and culture. NIU Journal of Humanities, 9, 69–77. [Google Scholar] [CrossRef]
  33. Panagiotidis, P. (2021, July 5–7). Podcasts in language learning. Research review and future perspectives. 13th International Conference on Education and New Learning Technologies (pp. 10708–10717), Online. [Google Scholar]
  34. Pandey, A. (2024). Podcasts in Kanauji: Assisting language teaching and revitalization. Language Documentation & Conservation, 18, 1–19. [Google Scholar]
  35. Păvăloaia, V.-D., Teodor, E.-M., Fotache, D., & Danileţ, M. (2019). Opinion mining on social media data: Sentiment analysis of user preferences. Sustainability, 11, 4459. [Google Scholar] [CrossRef]
  36. Rizwan, N., Deb, N., Roy, S., Solanki, V. S., Garimella, K., & Mukherjee, A. (2025). Dynamics of toxicity in political podcasts. arXiv, arXiv:2501.12640. [Google Scholar] [CrossRef]
  37. Royston, R. A. (2023). Podcasts and new orality in the African mediascape. New Media & Society, 25, 2455–2474. [Google Scholar]
  38. Salawu, A. (2006). Indigenous language media: A veritable tool for African language learning. Journal of Multicultural Discourses, 1, 86–95. [Google Scholar] [CrossRef]
  39. Selimovic, M., Abd Almisreb, A., & Amanzholova, S. (2024). The role of sentiment analysis in brand management and marketing: A comparative study. Procedia Computer Science, 251, 579–584. [Google Scholar] [CrossRef]
  40. Sharma, H. D., & Goyal, P. (2023). An analysis of sentiment: Methods, applications, and challenges. Engineering Proceedings, 59, 68. [Google Scholar] [CrossRef]
  41. Sharon, T. (2023). Peeling the pod: Towards a research agenda for podcast studies. Annals of the International Communication Association, 47, 324–337. [Google Scholar] [CrossRef]
  42. Tabassum, F., Raza, F., Khan, F. A., Akbar, S. B., Bangash, S. R., & Shahid, M. (2019). Investigating the impact of podcast learning in health via social network analysis. New knowledge in information systems and technologies (Vol. 3, pp. 493–499). Springer. [Google Scholar]
  43. Tobin, S. J., & Guadagno, R. E. (2022). Why people listen: Motivations and outcomes of podcast listening. PLoS ONE, 17, e0265806. [Google Scholar] [CrossRef] [PubMed]
  44. Uguru, J. O., & Okeke, C. O. (2020). On pronunciation in a multilingual dictionary: The case of lukumi, olukumi and Yoruba dictionary: De la prononciation dans un dictionnaire multilingue: Le cas du dictionaire lukumi, Olu kumi et Yoruba. Lexikos, 30, 519–539. [Google Scholar] [CrossRef]
  45. UNESCO. (2025). Endangered languages UNESCO classification. Available online: https://guides.lib.uconn.edu/c.php?g=1232158&p=9415488 (accessed on 7 October 2025).
  46. Washington, A. R. (2016). (Re) structuring structures, speech and selves: Yoruba language learning and redevelopment in Salvador da Bahia, Brazil. University of Pittsburgh. [Google Scholar]
  47. Yaman, I. (2016). The potential benefits of podcasts for language learning. Journal of Educational and Instructional Studies in the World, 6, 60–66. [Google Scholar]
Figure 1. Negative word clouds.
Figure 1. Negative word clouds.
Journalmedia 06 00179 g001
Figure 2. Neutral word clouds.
Figure 2. Neutral word clouds.
Journalmedia 06 00179 g002
Figure 3. Positive word clouds.
Figure 3. Positive word clouds.
Journalmedia 06 00179 g003
Table 1. Research objectives and methods.
Table 1. Research objectives and methods.
ObjectivesMethods
Objective1InterviewSentiment analysis
Objective 2Sentiment analysis
Objective 3InterviewSentiment analysis
Table 2. A Sample of Audience levels of Sentiments on the I Speak Yoruba Too Podcast.
Table 2. A Sample of Audience levels of Sentiments on the I Speak Yoruba Too Podcast.
S/NPositive_ReviewsNeutral_ReviewsNegative_Reviews
1Thank you, these are the best Yoruba lessons I have found online I actually found you on Audible too! Thanks so much! This lesson is too difficult
2This is excellent. Thank you so much What dialect of Yoruba is this?Only voice without us seeing the write top I don’t think is very helpful
3Amazing video keep it up Hello everyone, hope you enjoy this lesson. How many Yoruba Phrases do you know? Share in the comment section. The visualisation of this video is the best on your channel. Please, (1) prepare in advance, instead to do talk-show, (2) write on a screen about what you say, because visual memory is much stronger than audio (expecially for carriers of non-tonal languages)
4I would love to be fluent when I next visit my in-laws, Your lessons are excellent. When purchasing courses, do I have limited time access to the course, or do I have permanent access to the course? Thank you! Please correct the mistake in your answer for the last question to better guide us in learning
5These are very helpful!!! Thank you so much I thought I is Emi... What is Emi then? Not clear
Table 3. A sample of Audience feedback on Learn Yoruba Online.
Table 3. A sample of Audience feedback on Learn Yoruba Online.
S/NPositive_ReviewsNeutral_ReviewsNegative_Reviews
1Thanks so much sir, you are doing justice to the language, I am happy to find you, with you on my side I would speak Yoruba soon, thank you once more teacherMa se so ireti nu...e ma kaloThe audio is low
2You are the bestLike your pictureIt is not working for me.
3This was great! More of this subject pleaseWhat is the meaning when we, Yoruba, hold our left ear when we are angry? I have been noticing that oYou speak just too fast, I cannot hear what you are saying
4Greetings. Your classes are excellent. It would be very helpful if you could add acoustic tiles to your studio. The echo makes it hard to concentrate on your wonderful instruction.I am Kenyan. I hope to visit Nigeria soon. I will definitely reach out.The video keeps fading in and out. It is distracting and gave me a headache
5You are a great teacher!! You have a new student!! I will definitely contact you!Can we also say: Orun pa mi? When it is too hotAfter a video I don’t even know one word.
Table 4. Sentiment Analysis Summary.
Table 4. Sentiment Analysis Summary.
Sentiment Analysis Summary
SentimentNo. of sentimentScore of sentiment
Negative1550.7675
Neutral2240.7239
Positive3560.8993
Table 5. (a) Level of Sentiments Expressed in I Speak Yoruba Too Listeners’ Feedback. (b): Level of Sentiments Expressed in Learn Yoruba Online Listener Feedback.
Table 5. (a) Level of Sentiments Expressed in I Speak Yoruba Too Listeners’ Feedback. (b): Level of Sentiments Expressed in Learn Yoruba Online Listener Feedback.
(a)
S/NPositive_ReviewsNeutral_ReviewsNegative_Reviews
1These lessons are amazingDo you see value in this lesson? Is it helpful to you? Share your thoughts in the comment section. ThanksThe podcast lacked good content
2So thankful to have found your channel. I have been struggling trying to say my prayers and pronounce everything correctly. Your diction and articulation are perfect!I am starting from the beginning! I hope to catch up soon! Thank you! Repetitive content
3Nice! Now I know some words! Good job!I have been on my spiritual journey and have realised that many of my roots are in Yoruba. And it is surprising how fast and easy this language is coming to me. I found your channel on Spotify. Disappointed with the quality and delivery of the podcast
4Thank you, these are the best Yoruba lessons I have found online I subscribed to this channel today, ready to learn everything that has been taught here from day one, as I will be watching all the videos and taking notes as well. I hope to improve my Yoruba speaking skills. Thank you for doing thisThere was no clear direction in the discussion
(b)
S/NPositive_ReviewsNeutral_ReviewsNegative_Reviews
1You are too much! We appreciate God bless you.I met a friend who is Nigerian and speaks Yoruba and I am so excited to surprise him with some phrasesWhy does the teacher look angry
2Great. Thanks. I love it. Please record it a little louder.I really learnt a lot in this videonot helpful sorry
3This is amazing, thanks a lotAre you in Slovakia?I don’t understand
4I love this video...I want Tide to call me = Mo fè ki Tide pé miYoruuba is so difficult
5Gradually you are becoming the Best Yoruba Language Teaching Channel on YouTube. Respect from Afghanistan.YorubaAn example is not enough
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Ajala, B.C.; Salawu, A.; Fadipe, I.A.; Aromavo, Y.P. Exploring the Nexus of Opportunities and Challenges in Indigenous Language Podcasting Through Natural Language Processing of User-Generated Content. Journal. Media 2025, 6, 179. https://doi.org/10.3390/journalmedia6040179

AMA Style

Ajala BC, Salawu A, Fadipe IA, Aromavo YP. Exploring the Nexus of Opportunities and Challenges in Indigenous Language Podcasting Through Natural Language Processing of User-Generated Content. Journalism and Media. 2025; 6(4):179. https://doi.org/10.3390/journalmedia6040179

Chicago/Turabian Style

Ajala, Bukola Christiana, Abiodun Salawu, Israel Ayinla Fadipe, and Yetunde Pesu Aromavo. 2025. "Exploring the Nexus of Opportunities and Challenges in Indigenous Language Podcasting Through Natural Language Processing of User-Generated Content" Journalism and Media 6, no. 4: 179. https://doi.org/10.3390/journalmedia6040179

APA Style

Ajala, B. C., Salawu, A., Fadipe, I. A., & Aromavo, Y. P. (2025). Exploring the Nexus of Opportunities and Challenges in Indigenous Language Podcasting Through Natural Language Processing of User-Generated Content. Journalism and Media, 6(4), 179. https://doi.org/10.3390/journalmedia6040179

Article Metrics

Back to TopTop