Transcribing and auto-generating meeting summaries
The challenge
Across the public sector, millions of public servants deliver critical services through meetings, interviews and appointments. They then spend hours processing and summarising the outputs of these conversations, filling out forms, writing briefings or crafting official minutes. In some cases, dedicated minute takers are used to support this process. However, it is not possible for every public sector interaction to have a dedicated minute taker. This means that public servants are often required to do two jobs at once: be fully present in the conversation while documenting all key points of the interaction, ready to feed into a host of structured formats afterwards.
The solution
Using a transcription engine and a large language model (LLM), Minute produces bespoke summaries in the formats that public servants really need. Users have the choice between uploading existing audio and video files, or can record live directly in the app.
The results
In early user testing in central government, Minute reduced the time taken to complete minutes for a 60 minute meeting by one hour, which users reallocated to more critical tasks. Full Quality Assurance and publication of this analysis is forthcoming.
Nearly 90% of users described minuting as the one of the less enjoyable aspects of their job, with nearly half describing it as the 'least enjoyable'.
Learnings / insights
There is a strong need for AI transcription tooling across the public sector.
There are common templates required across councils, including the statutory requirement to publish official minutes.
Working directly with a community of enthusiastic users at an early development stage helps to make important changes early. Local Authorities have strong desire and capability to contribute to these conversations.
Get access to Minute
Minute is not available to new departments or local authorities.
It's being pilot tested with 22 local authorities and some central government departments. We're working with MHCLG to scale it for all local authorities, focusing on frontline services.
Details
Organisation name
Department for Science, Innovation and Technology (DSIT) / Incubator for Artificial Intelligence (i.AI)
Minute
Transcribing and auto-generating meeting summaries
The challenge
Across the public sector, millions of public servants deliver critical services through meetings, interviews and appointments. They then spend hours processing and summarising the outputs of these conversations, filling out forms, writing briefings or crafting official minutes. In some cases, dedicated minute takers are used to support this process. However, it is not possible for every public sector interaction to have a dedicated minute taker. This means that public servants are often required to do two jobs at once: be fully present in the conversation while documenting all key points of the interaction, ready to feed into a host of structured formats afterwards.
The solution
Using a transcription engine and a large language model (LLM), Minute produces bespoke summaries in the formats that public servants really need. Users have the choice between uploading existing audio and video files, or can record live directly in the app.
The results
Learnings / insights
Get access to Minute
Minute is not available to new departments or local authorities.
It's being pilot tested with 22 local authorities and some central government departments. We're working with MHCLG to scale it for all local authorities, focusing on frontline services.
Details
Links