All use cases
// Mobile App Development

Voice assistant app

An app people talk to, and that talks back.

Engagement that fits
Fixed-scope build, Product partnership
Stack
React NativeTypeScriptOpenAI APIClaude APINestJS

// Overview

Where this product fits

// Audience

Who it is for

Products whose users have their hands or eyes busy, such as drivers, field workers, cooks or learners, and would rather speak to the app than type into it.

// Problem

The problem

Typing on a phone is slow on the move, and a voice app that lags, mishears or loses the thread of a conversation gets closed for good.

// Deliverables

What we would build

  • Voice input with live transcription

    Users speak instead of typing, and their words appear on screen as they talk, so they can see the app understood them.

  • Spoken replies from the assistant

    The assistant answers out loud in a natural voice and shows the same reply as text, so users can listen without looking at the screen.

  • An assistant grounded in your content

    Answers come from your own guides, catalogue or courses rather than the open web, so the assistant stays on topic and in your voice.

  • A text fallback for quiet places

    When speaking is not an option, users type their question and get the same answers in the same conversation, without losing context.

  • Conversation history per user

    Each user's past conversations are saved to their account, so they can pick up where they left off on any of their devices.

  • App Store and Google Play releases

    Both apps are prepared, submitted and released through the App Store and Google Play, including the microphone permissions each store reviews.

// Process

How it works

  1. 01.

    The app records and streams speech

    Audio is captured on the phone and streamed as the user speaks, so transcription starts before they finish their sentence.

  2. 02.

    Speech becomes text the model reads

    A speech recognition model turns the audio into text, which the assistant reads together with the conversation so far.

  3. 03.

    The model answers from your content

    The assistant looks up the relevant parts of your content and writes a reply grounded in them, not in guesswork.

  4. 04.

    The reply is spoken and shown on screen

    The answer is turned into speech and played back, while the same words appear on screen for users who prefer to read.

What's next

Let's build
your product.

Tell us what you're building and where it stands. We'll scope it with you on a call and give you a straight answer on fit.

  1. 01

    Book a call

    Pick a time on our calendar and tell us what you want to build.

  2. 02

    Shape the scope

    Together we name the users, the first release and what success looks like.

  3. 03

    Get a proposal

    We send a written plan with the scope, the steps and the engagement that fits.

  4. 04

    Start the work

    We begin with Discovery and show you working software early and often.