Applications are open for the Sahara CodeSwitch Africa Challenge 2026, an initiative inviting developers, startups, researchers, product teams, students, and other builders across Africa and the African diaspora. Participants are challenged to address significant unsolved problems in voice AI by using Sahara’s code-switching APIs to build and benchmark real-world applications. The challenge aims to demonstrate how effectively current speech models handle code-switched audio in practical, real-world settings.
Challenge Task
• The primary objective is to create or improve a real-world voice AI application.
• The transcript generated from voice input must be used to perform a downstream task rather than simply converting speech to text. Examples include a voice assistant for completing bookings, a voice-driven triage tool, a spoken-language form filler, or a voice interface that performs an action on a user’s behalf.
Challenge Categories
Submissions are welcomed across five tracks designed to address real-world challenges and opportunities in Africa:
• Health: Clinical documentation, patient intake, community health, maternal health, public health, hospital workflows, and frontline care.
• Legal & Public Services: Court transcription, legal aid, citizen services, civic reporting, complaints, access to justice, and public-sector workflows.
• Fintech, Telco & Customer Experience: Voice banking, KYC, payments, collections, customer support, call centers, agent assistance, and customer insights.
• Agriculture & Education: Farmer advisory services, agricultural extension services, tutoring, classroom learning, local-language education, and skills training.
• Other High-Impact Use Cases: Humanitarian response, accessibility, media, religious services, commerce, transport, culture, and other meaningful applications of code-switching voice AI.
Prizes and Incentives
• A total cash prize pool of $10,000 will be awarded to top-performing solutions across the challenge categories.
• Top teams will also receive additional benefits, including:
• Partner Demo Day: Opportunities for selected teams to showcase their solutions.
• Greater Visibility: Exposure to sponsors, judges, funders, and ecosystem partners.
• Pilot Opportunities: Potential opportunities for pilot projects or deployment discussions.
• Sahara API Credits: Additional API resources to support continued product development.
• Recognition: Promotion through Intron and partner channels.
Eligibility Criteria
• The challenge is open to builders across Africa and the African diaspora, with participants able to apply individually or as part of a team.
• Participation is especially encouraged from:
• Developers and engineers
• AI/ML researchers
• Startups and product teams
• Student teams
• Open-source builders
• NGOs and civic-tech teams
• Companies with voice-based or customer-facing workflows
Application Requirements
Each submission must meet the following requirements:
• Use voice input to perform a downstream, agentic task rather than simply converting speech to text.
• Benchmark performance on code-switched audio across at least three speech models. This must include an Intron Sahara API and at least two other models of the team’s choice, which may be global, local, open-source, commercial, pretrained, or fine-tuned models.
• Target a specific vertical or use case within one of the designated challenge categories.
• Where possible and permitted, submit the code-switched audio samples used for testing along with basic metadata, such as language pair, domain, accent or country, device type, and noise conditions.
Application Details
• Deadline for Submission: September 15, 2026 (16 September 2026, 00:59 GMT+1)
• To Apply: Visit the official application page
• For More Information: Visit the Sahara CodeSwitch Africa Challenge website
The Sahara CodeSwitch Africa Challenge 2026 offers developers, researchers, startups, and other innovators an opportunity to advance voice AI for Africa’s multilingual and code-switching environments. By developing practical applications and benchmarking speech models under real-world conditions, participants can contribute to more effective and inclusive voice technologies while competing for cash prizes, visibility, API credits, and potential pilot opportunities.