Pioneering Voice AI Solutions for Overlooked Markets: The Journey of AethexAI
In the rapidly evolving landscape of artificial intelligence, voice AI has emerged as a transformative force in customer support and service sectors. However, the development of voice AI solutions that deliver human-like interactions with minimal latency presents unique challenges, particularly in regions such as Africa and the Middle East. Recognizing this gap, AethexAI, a startup founded in 2025, is dedicated to crafting voice AI technologies tailored to these underserved markets.
Founding Vision and Leadership
AethexAI was established by Mariama Diallo and Ayooluwa Odemuyiwa, professionals with impressive backgrounds in finance and technology. Diallo, serving as CEO, previously held positions at Goldman Sachs and later contributed to the growth of YC-backed ModelML. Odemuyiwa, the CTO, is a Caltech graduate who worked at Meta and pursued further education at Stanford Business School before co-founding AethexAI. Their shared ambition to create impactful solutions for emerging markets led them to explore opportunities in voice AI.
Addressing Regional Challenges
The adoption of AI tools for automating business operations is a global trend. However, in regions like Africa and the Middle East, this transition has encountered significant obstacles. For instance, an Egyptian call center’s attempt to automate a substantial portion of its calls resulted in subpar outcomes, leading to the abandonment of the system. Additionally, numerous support centers across Africa have reported difficulties in recruiting engineers capable of implementing cost-effective call automation solutions.
Odemuyiwa highlighted the technical hurdles faced in these regions:
The latency and jitter that we saw on automated calls in this region were outrageous. If we had become orchestrators, we might have had to use large models that were hosted outside the region, resulting in higher latency. We realized that in order for this to work, we have to use very small models and cut latency at every step.
Innovative Technological Approach
To effectively serve their target markets, AethexAI made the strategic decision to develop proprietary models and an orchestration layer from the ground up. This approach was essential to accommodate the diverse dialects of English, French, and Arabic prevalent in the regions they aim to serve. By focusing on smaller models, AethexAI successfully addressed latency issues while maintaining high accuracy levels. Their Kora series of models, ranging from 300 million to 1.7 billion parameters, exemplifies this commitment to efficiency and performance.
Data Acquisition and Model Training
Training effective AI models necessitates substantial and relevant data. AethexAI adopted innovative methods to gather the necessary datasets:
– Anonymized Call Recordings: Collaborating with a call center partner, AethexAI utilized anonymized recordings to train their models, ensuring privacy and compliance with data protection standards.
– Radio Station Partnerships: The company distributed hard drives to radio stations across Africa to collect a diverse range of audio data, capturing various accents and speech patterns.
– Contributor Network: To annotate data and accurately pronounce local names, AethexAI built a network of university students, fostering community involvement and reducing operational costs.
These efforts have enabled AethexAI to manage over 17,000 calls daily, demonstrating the scalability and effectiveness of their solutions.
Client-Centric Business Strategy
Understanding that many clients are new to voice AI technologies, AethexAI emphasizes a consultative approach:
We always tell customers that we cannot be everything for everybody right now. We’re small. When we start talking to a company, we ask them to pick one use case that is the most important to them to start [with], Diallo explained.
This strategy allows AethexAI to deliver tailored solutions that address specific client needs, ensuring successful implementation and measurable outcomes.
Diverse Industry Applications
While open to various industries, AethexAI currently focuses on applications such as:
– Debt Collection: Automating outbound calls to streamline the debt recovery process.
– Customer Activation: Engaging potential customers through personalized voice interactions to enhance conversion rates.
– Know Your Customer (KYC) Verification: Facilitating identity verification processes for banks and telecom companies, ensuring compliance and security.
To support these applications, AethexAI is hiring forward-deployed engineers on a contract basis to serve local markets and is establishing channel partnerships with telecom providers to manage telephony for voice AI calls.
Market Differentiation and Competitive Edge
Walter Badoo, co-founder and managing partner of 4DX Ventures, emphasized the unique challenges and opportunities in the Africa and Middle East markets:
Enterprises in Africa and the Middle East process roughly three times the call volume of their Western counterparts, as voice is still the dominant channel for customer interaction. Incumbent systems were built for Western markets characterized by high-end GPU infrastructure, standard English and European speech environments, and enterprise workflows common in the US and Europe. That creates real gaps when enterprises need systems that handle dialects, code-switching, and informal speech patterns, and that work within their existing telephony infrastructure and their actual price points.
This insight underscores the necessity for specialized solutions like those offered by AethexAI, which are designed to meet the distinct needs of these regions.
Conclusion
AethexAI’s innovative approach to voice AI technology, combined with a deep understanding of the unique challenges in Africa and the Middle East, positions the company as a trailblazer in the field. By developing tailored solutions that prioritize low latency, accuracy, and cultural relevance, AethexAI is not only addressing a significant market gap but also paving the way for enhanced customer interactions in regions that have long been overlooked by mainstream AI developments.
