Hasab AI Launches Multilingual Meeting Note-Taking Tool for Ethiopian Languages

Seven months after its official debut, Ethiopian speech-recognition startup Hasab AI has introduced a multilingual, AI-powered meeting note-taking feature, targeting the failure of global platforms to handle low-resource languages. 

The new feature supports Amharic, Afaan Oromo, and English, and integrates directly with Google Meet, Microsoft Teams, Zoom, and Webex. Users can deploy the tool by allowing it to join meetings live or by uploading recorded audio or video files for processing. The system generates real-time transcripts with speaker attribution and timestamps, produces structured summaries in English, and delivers action points and key insights. An embedded AI chatbot allows users to query meeting notes, turning recorded discussions into searchable, decision-ready material. 

“Calendar integrations allow the tool to sync with Google and Microsoft calendars, automatically joining scheduled meetings based on user-defined permissions,” said Kidus Yared, founder and CEO of Hasab AI.

Accessible via meet.hasab.ai, the platform can automatically email participants transcripts and summaries after each meeting, a feature that can be enabled or disabled. 

“All recordings and outputs are encrypted in transit and at rest, with no third-party data sharing and full user control over data deletion,” Kidus told Shega. 

Hasab AI is positioning the product primarily for enterprises, startups, and government organizations that rely heavily on meetings, while also making it available to small teams. 

The service is offered through a regular tier for small teams and a premium enterprise tier for teams of more than 10 people. The premium tier adds team management, role organization, dashboards, analytics, and reporting features. A 15-day free trial offers limited daily meeting hours; after which, users can subscribe starting at $20 per month, payable via local or international payment methods. 

“Most existing meeting tools fall short when it comes to accurately capturing and summarizing meetings conducted in local languages,” Kidus told Shega, noting that this limitation is particularly acute in Ethiopia’s multilingual workplaces. He described the new feature not only as a productivity tool, but as foundational infrastructure for building more advanced local-language AI systems in the future.

Ethiopian workplaces often operate in a deeply mixed linguistic register, with Amharic and English blended within the same sentences rather than used side by side. Kidus said Hasab AI’s models can handle this kind of language mixing. But he acknowledges that heavily interwoven speeches remain difficult to transcribe accurately. Improving performance in such highly mixed-language meetings, he added, is an active and ongoing area of development.

A handful of business and financial institutions have already adopted Hasab’s meeting note-taking features, including Globe Dock Academy, GoodayOn, and Ring Cloud. The company is also in ongoing discussions with government agencies and financial institutions and is working with the Ethiopian AI Institute.

Kirubel Akalu, founder and CEO of edtech startup Globe Dock Academy, said  Hasab's integration with RingCloud, their cloud telecom provider, has been very helpful. "As RingCloud subscribers, we now access agent call recordings and transcriptions in one CRM dashboard."

Since launch, Hasab AI says it has attracted around 10,000 users. The startup is part of the NVIDIA Inception program, which provides technical resources to help scale its audio intelligence platform for African and other underserved languages. In addition, Hasab AI was selected among 20 “Compute Ready” companies in the AI Hub for Sustainable Development program. Kidus said participation in these programs is helping the company address two major challenges: access to computing resources and technical support.

Hasab AI has also released a free Telegram bot that converts voice notes and audio files into text in supported languages, including Amharic, English, Tigrinya, and Afaan Oromo.