12 bilingual models
Sahara v2.5 expands bilingual speech intelligence with 12 models built for conversations that naturally move between languages.
Our most intelligent voice AI model yet — built to understand how Africa actually speaks, including switching between languages mid-sentence.
Choose a workflow and play the voicebot demo.
Ready to help.
Sahara v2.5 expands bilingual speech intelligence with 12 models built for conversations that naturally move between languages.
Sahara v2.5 can follow one conversation as it moves naturally across English, Kinyarwanda and French without resetting context.
Sahara v2.5 extends from understanding African speech to generating natural voice responses across 13 supported languages.
Seven additional languages expand Sahara v2.5 from 56 to 63 total languages, bringing African voice AI to more speakers, markets and real-world use cases.
Sahara v2.5 improves recognition quality for real-world African speech, helping voice systems understand conversations more reliably across practical environments.
From collections and healthcare to courtrooms and clinical documentation, Sahara is already delivering measurable outcomes across real African workflows.
BRANCH
Sahara-powered Branch collections agents recovered more than ₦1.2 million in delinquent loans in one week, with record after-hours and weekend repayments, outperforming human agents on delinquent loans over 356 days.
BRANCH
Collaborating with Intron to build a Branch-aligned collections bot was a rewarding experience. Customers engaged naturally even after hours and on weekends. Conversations were human and effective at delivering real payments on delinquent loans, sometimes over 2 years old. Adding local language and language-mixing support in this v2.5 release helps us reach even more customers.
AUDERE
Youths across South Africa can now speak naturally with Audere’s Self-Cav reproductive health WhatsApp chatbot using voice to express themselves freely instead of being confined to text chat.
ARM
Using Intron AI models, we’ve seen significant improvement in transcription and summaries compared to models we previously explored. Their systems capture context and nuance better, leading to more accurate results.
OGUN STATEOver one year working with Intron, deployment has grown from an initial 3 pilot courts to 9 courts, with the goal of bringing all 18 courts across the state on board. Yoruba is frequently spoken during testimony, and Sahara v2.5 helps ensure those segments are properly captured in court records.
NAIROBIA 14-minute Swahili-English doctor-patient consultation in Nairobi is converted into a structured clinical note in less than 30 seconds, reducing documentation burden so physicians can focus more on patient care.
PAMOInternet connectivity had blocked the adoption of many transformative solutions. Intron’s offline medical scribe deployment has transformed clinical operations. At over 2,000 consultations using voice AI and a 4× documentation speed improvement, the team is now benefiting from AI without depending on reliable internet.
GOOEY.AI
On 4 of 6 Nigerian languages, independent Medical Question Answering benchmarks for the Gates Foundation and CLEAR Global showed Intron to be on par with or better than transcription alternatives including Gemini and Meta’s Omnilingual LLM. Sahara v2.5 is one of the first dedicated models for African code-switching they have evaluated.
Sahara-v2.5 — the voice AI model that makes these bots work where others fail.
Follows every word when callers switch mid-sentence without losing intent or resetting the transcript state.
Natural voice generation that sounds like someone from your customer's region — regional inflections, proper pacing.
Optimized on noisy, low-bandwidth telephone lines, performing 20%+ better on typical market background noise.
Outperforms standard systems by 64% on local name recognition and 35% on alphanumeric values like account numbers and bank codes.
Executive Outcome Bar
Collections
Automate repayment reminders, promise-to-pay capture, rescheduling and escalation in local languages — including after hours and weekends.
Contact center
Auto-resolve routine enquiries, summarise calls, route complex issues to the right agent and reduce repetitive work across your team.
Growth
Qualify leads, educate customers, reactivate dormant users and surface next-best offers through conversations that already happen.
Quality & compliance
Move from small-sample QA to 100% call transcription, scoring, supervision and automated compliance flagging.
Financial access
Let customers bank by speaking naturally across the channels they already use.
Trust & identity
Apply voice authentication and biometric verification to your highest-risk financial workflows.
See how Intron voice bots perform under pressure
An Intron autonomous debt-collection agent recovered over N1.2M in delinquent loans in one week and delivered the highest number of weekend payments the lender had ever recorded.
On delinquent loans over 365 days, Intron's autonomous agent outperformed human agents by more than 4x.
In a May 2026 Accion International study, some respondents didn't realise they were speaking with a bot.
Gates Foundation-commissioned benchmarking: Sahara outperformed on 5 of 7 Nigerian languages vs. Gemini and Meta ASR.
100% contact-center transcription and analysis, giving full visibility into compliance, churn, agent performance, and customer experience.
In use or in pipeline with Branch, ARM, Carbon, Sycamore, Regxta, Huawei AICC, Union Bank, Stanbic IBTC, Sterling, Interswitch, Computech, Pngme, and ABSA Kenya/Uganda.
40+ partners across 6 countries, $36K MRR, 4x YoY revenue growth, and 6x month-over-month usage growth.
Recover more without adding more agents.
Intron voice agents reach borrowers consistently with localised, human-like repayment conversations. Automate reminders, promise-to-pay capture, rescheduling, escalation, and after-hours calling across English, Pidgin, Hausa, Yoruba, Igbo, Swahili, and more, including customers who switch between two of them in the same call.
Reduce wait time. Resolve routine issues automatically.
Move from rigid IVR to intelligent voice self-service. Customers check balances, report failed transfers, block cards, request statements, ask repayment questions, and get app support. Complex cases escalate to your agents with the transcript, summary, and customer intent already attached.
Turn customer conversations into revenue.
Voice agents qualify leads, explain products, follow up on abandoned applications, reactivate dormant customers, and route high-intent prospects to your team. Call summaries, sentiment, intent, and next-best actions flow back into your CRM.
Make onboarding easier for every customer.
Customers complete KYC, merchant forms, loan applications, and account opening by speaking instead of typing. Intron captures names, addresses, dates, IDs, income details, repayment preferences, and business information accurately, with structured outputs your downstream systems can use.
Analyse every call. Coach every agent. Catch risk earlier.
Every customer conversation becomes structured intelligence. Intron transcribes, summarises, scores, and analyses calls for compliance gaps, weak verification, fraud risk, churn signals, customer sentiment, and agent performance.
Let customers bank by speaking naturally.
Voice commands for transfers, bill payments, statements, card blocks, loan applications, support, and merchant workflows. Deploy across mobile apps, web, IVRs, ATMs, kiosks, field-agent tools, and low-connectivity environments.
Follows a caller who switches language mid-sentence, across 12 African languages
Breaks or mistranscribes
Not supported
Fluent TTS in 12 African languages
Foreign cadence, few languages
Text-first or robotic
Built for 500+ accent variations
Limited coverage
Usually English-biased
Tuned for market noise, generators, overlapping speakers
Often brittle
Assumes quiet input
Optimised for African names, IDs, currencies, dates
General-purpose
Not specialised
Hands to a human with transcript, summary, and intent attached
Varies
Ticket only
API, cloud, telephony, CRM, WhatsApp, IVR, offline options
Often cloud only
SaaS-only
Every call transcribed, scored, and analysed
Not workflow-native
Limited analytics
12 African languages, code-switching, native TTS, and noise-resilient ASR – all accessible via a unified simple API with low-latency endpoints.
Build with Sahara-v2.5 →Ọpọlọpọ eniyan tọ́ wá sí ọjà náà lónìí...
Detected: YorubaMany people came to the market today to buy goods.
Detected: EnglishMutane da yawa sun zo kasuwa yau...
Detected: HausaChoose the type of enquiry that best fits what you need. We’ll route your submission to the right team.
Gain deep insights into the evolving technological landscape with our upcoming 2026 Africa Voice AI Report. The report explores the trends, challenges, and opportunities shaping the future of speech interfaces across the continent.