🏠 Overview › 🎯 Features ›🎯 Speech-to-Text Input for Reports

Speech-to-Text Input for Reports

Feature Detail

medium complexity extracted Activity Registration Confidence: 100%
2
Components
58
Shared
0
User Stories
Yes
Analyzed

Description

Provides a speech-to-text input widget enabling peer mentors to dictate activity reports instead of typing on a mobile keyboard. Designed exclusively for post-activity report writing — never for recording during conversations, which workshop participants flagged as a serious privacy concern. The widget integrates into free-text fields across the activity wizard and structured report forms, supporting peer mentors with motor impairments, visual disabilities, or low typing proficiency on both iOS and Android platforms.

User Flow

Speech-to-Text Input for Reports user flow
Click to expand

Analysis

Business Value

For peer mentors with motor impairments, visual disabilities, or limited digital literacy, typing on a mobile keyboard is a significant barrier to complete and accurate reporting. Blindeforbundet and HLF both identified speech-to-text as an accessibility priority that increases report quality and length. Removing the typing barrier also encourages more detailed narrative entries, which improve the usefulness of activity records for coordinator review and organizational memory. This feature directly supports the WCAG 2.2 AA accessibility mandate and extends genuine usability to the broadest possible range of peer mentor profiles, including older volunteers who represent a significant portion of the user base.

Implementation Notes

Implemented using the Flutter speech_to_text plugin, which delegates to native platform APIs (iOS SpeechRecognizer, Android SpeechRecognizer) for on-device transcription with no audio storage or transmission. SpeechInputWidget is a reusable component wrapping any TextField with a microphone button, displaying live transcription and allowing correction before insertion. SpeechRecognitionService manages permission state (microphone), language selection (Norwegian Bokmål default), error recovery for recognition failures, and graceful degradation when the device lacks speech support. AudioProcessingInfrastructure configures the platform audio session to prevent conflicts with calls or media. WCAG compliance requires that the microphone button has a descriptive accessible label and that activation/deactivation state is announced to screen readers via Semantics.

Components (60)

Service Layer (1)

service Speech Recognition Service medium

Infrastructure (1)

infrastructure Audio Processing Infrastructure low

Shared Components

These components are reused across multiple features

Infrastructure (20)

infrastructure JWT Infrastructure medium Shared infrastructure JWT Infrastructure low Shared infrastructure Auth Middleware medium Shared infrastructure Report API Middleware low Shared infrastructure Report Access Guard low Shared infrastructure Approval Audit Logger low Shared infrastructure End-to-End Encryption Infrastructure high Shared infrastructure Email Delivery Infrastructure low Shared infrastructure Invitation Token Infrastructure medium Shared infrastructure Signed URL Provider low Shared infrastructure Multi-Tenancy Infrastructure high Shared infrastructure Calculator Export Infrastructure medium Shared infrastructure Request & Response Interceptors high Shared infrastructure Design Token System medium Shared infrastructure Semantic Labels Infrastructure medium Shared infrastructure Connectivity Infrastructure low Shared infrastructure Local Database Infrastructure high Shared infrastructure Secure Storage Infrastructure medium Shared infrastructure Connection Pool Infrastructure medium Shared infrastructure Database Migration Infrastructure medium Shared

User Stories

No user stories have been generated for this feature yet.