wandel.ai

A Speech To Text iOS app and platform that turns spoken words into beautiful notes

See more

The Brief

Build the Swiss Army knife of note taking, turning hyper-accurate transcripts into well-structured notes with the help of AI. Stand out with reliable, accurate and privacy-first processing of voice recordings.

The Brief

5x Faster Notes

Taking notes can be difficult in many situations. Thoughts don't always wait for a calm moment with a keyboard at hand. Usage of LLMs for transcription and text conversion can significantly lower the time it takes to create notes as well as open up new ways of bringing ideas into the digital world.

5x Faster Notes

Build for Teams

There are many applications of capturing feedback and information on the ground, and for teams. We wanted to make sure that Wandel's architecture supports the needs of teams from the get-go, and plan to iteratively adapt the interface to that architecture.

Build for Teams

The Serverless Architecture

We built a fully serverless backend on AWS that will scale with use on demand. Connected to the backend is a native iOS app written in Swift, with a normalized local cache, kept in sync with the cloud infrastructure to maximise on-device availability and usability. Additionally we built a web admin dashboard that allows us to derive insights and manage system data more easily. We offload operational complexity, and ensure our backend can scale with unknown demand by choosing a serverless approach to building our platform. All of this while having a pay per use cost structure for the backend, so we’re not paying for resources that aren’t being used.

The Serverless Architecture

Technical Highlights

Fully managed GraphQL API via AWS AppSync including real time updates

A normalised cache in a local SQLite database on the client

An LLM integration service allowing us to easily plug in several LLM providers such as Deepgram for transcription and OpenAI and Anthropic for text conversion.

DynamoDB implementing Single Table Design for performance and cost

Event Driven architecture with Amazon SNS and Amazon EventBridge

Data Lake implemented with Kinesis, AWS S3 and Amazon Athena

Defined using Cloud Development Kit (CDK), allowing consistent deployments and flexibility

The App

We designed and developed a native iOS speech to text app. The robust backend implementation and intuitive frontend design enables us to capture speech with high accuracy and then flexibly convert this speech to any kind of text the user wants.

The App

App Store

Wandel.ai

An intuitive recording experience

The recording interface is simple and clean, making it easy for the user to focus on speaking their mind. Recording state is carefully managed to provide relevant feedback at any step in the process.

Multi-language support

The Wandel AI can transcribe in 15+ languages, so we made sure that the interface matches it's capabilities. Every word in the app is dynamically localized to multiple languages.

Multi-language support

The Wandel AI can transcribe in 15+ languages, so we made sure that the interface matches it's capabilities. Every word in the app is dynamically localized to multiple languages.

Intuitive Document Editing

Editing notes and transcripts is handled in a performant document view. Users have full control over their notes and can edit text themselves or tap into the helping hands of AI.