Draft:Zabbot
Submission declined on 5 October 2025 by Bakhtar40 (talk).
Where to get help
How to improve a draft
You can also browse Wikipedia:Featured articles and Wikipedia:Good articles to find examples of Wikipedia's best writing on topics similar to your proposed article. Improving your odds of a speedy review To improve your odds of a faster review, tag your draft with relevant WikiProject tags using the button below. This will let reviewers know a new draft has been submitted in their area of interest. For instance, if you wrote about a female astronomer, you would want to add the Biography, Astronomy, and Women scientists tags. Editor resources
|  |
Comment: In accordance with the Wikimedia Foundation's Terms of Use, I disclose that I have been paid by my employer for my contributions to this article. BolaAg (talk) 23:44, 4 October 2025 (UTC)
 | |
.png) Screenshot of Zabbot on October 4, 2025 | |
Type of site | Language learning, Artificial Intelligence, Education Technology |
|---|---|
| Founded | May 5, 2025 |
| Headquarters | Ithaca, NY |
| Area served | Worldwide |
| Founder(s) | Bola Agbonile |
| URL | zabbot.com |
| Registration | Yes |
| Launched | 2025 (Beta) |
| Current status | Active |
Zabbot is a language learning platform that uses AI speech recognition and text-to-speech technologies to help learners master pronunciation, tone, and grammar in heritage languages, beginning with Yorùbá. Unlike conventional language apps, Zabbot integrates cultural elements such as proverbs, folktales, and idiomatic expressions to connect learners to the social and emotional depth of the language.
History
[edit]Zabbot was founded in May 2025 by Bola Agbonile [1], an American-based technologist, product strategist, and AI innovator known for her work in digital customer experience and language technology. A U.S. patent holder [2] with over two decades in engineering and product leadership in Nigeria and USA. The Zabbot platform was inspired by Agbonile’s personal mission to teach Yorùbá to her adult children and preserve linguistic heritage within diaspora families.
Development of the Zabbot web app began in July 2025 with early user testing and community feedback sessions [3]. The platform’s beta program launched in August 2025, attracting over 150 testers across multiple countries .
Features
[edit]- Pronounce with Pàrà: A pronunciation feedback tool powered by dual AI models (Whisper ONNX and SpeechBrain ECAPA) that evaluates tone, clarity, and rhythm.
- Listen with Òwè: A text-to-speech storytelling feature for hearing Yorùbá phrases, proverbs, and short stories.
- Conversational Tutor: An AI-driven dialogue assistant that lets learners practice speaking Yorùbá in real time.
- Cultural Immersion: Courses include cultural notes, traditional stories, and idiomatic expressions.
- Gamified Learning: Learners earn badges such as 7-Day Streak), Vocab Master), and Culture Keeper.
- Marketplace: Connects learners with native tutors, cultural artisans, and vendors offering relevant inspiring products and services.
Technology
[edit]Zabbot integrates multiple artificial intelligence engines, speech APIs, and runtime frameworks to support its conversational and pronunciation learning features.
- Chat with Òré: Powered by OpenAI’s Chat Completions API (https://api.openai.com/v1/chat/completions), this feature enables natural-language dialogue between learners and an AI tutor that provides explanations, translations, and cultural insights.
- Pronounce with Pàrà: Utilizes a combination of machine learning and audio signal processing tools for pronunciation analysis and tone correction, including:
- Whisper ONNX Encoder — runs locally via ONNX Runtime for audio-to-embedding inference and speech comparison.
- File System API — reads and writes user audio files during pronunciation feedback.
- Path API — manages local file paths for audio storage and retrieval.
- WAV Parser API — decodes and processes
.wavaudio files before ML inference. - FFT (Fast Fourier Transform) Engine — extracts key acoustic features such as pitch, tone, and spectral energy.
- TensorFlow.js Runtime — executes TensorFlow models in Node.js for tone scoring, similarity analysis, and adaptive feedback.
- Listen with Òwè: Integrates with the Spitch Speech API and Spitch Diacritics API (https://api.spi-tch.com/v1/speech, https://api.spi-tch.com/v1/diacritics) to deliver high-quality Yorùbá text-to-speech playback with tonal accuracy and diacritic precision.
Zabbot’s modular architecture allows future integration with additional AI speech services and linguistic APIs to extend support to other African and tonal languages.
Funding and Development
[edit]Zabbot is bootstrapped by its founder, with grant and accelerator applications under review. It has engaged early partners in AI language modeling and cultural preservation and plans to expand to other underrepresented languages starting in 2026.
See also
[edit]References
[edit]- ^ "Zabbot launches Beta version of AI-powered Yoruba learning platform". LinkedIn. 2025-08-10. Retrieved 2025-09-04.
- ^ "US12026413B1 : Machine-learning system and print-queue based estimator for predicting wait times". US Patents. 2025-08-10. Retrieved 2024-02-27.
- ^ "Zabbot Pitch to Investors".
- in-depth (not just brief mentions about the subject or routine announcements)
- reliable
- secondary
- strictly independent of the subject
Make sure you add references that meet all four of these criteria before resubmitting. Learn about mistakes to avoid when addressing this issue. If no additional references exist, the subject is not suitable for Wikipedia.