---

name: ios-speech-recognition version: 0.1.0 author: cmtzco description: iOS Speech Framework integration for voice-to-text transcription. When to use: when implementing speech recognition, building dictation features, adding voice commands, converting audio to text, or working with SFSpeechRecognizer. What problems it solves: provides on-device speech recognition with high accuracy, handles real-time audio transcription, manages microphone permissions, and processes speech results efficiently.

iOS Speech Recognition

Overview

The Speech Framework enables iOS apps to recognize spoken words and convert them to text. This skill covers setting up SFSpeechRecognizer, configuring AVAudioEngine for real-time audio capture, handling authorization, and processing recognition results.

Identified Patterns

The following patterns were identified in this codebase that this skill addresses:

Pattern 1: Speech Recognizer Setup

• Location: DadBrain/DadBrain/ViewModels/VoiceRecorder.swift
• Description: SFSpeechRecognizer initialization with locale configuration
• Frequency: 1 complete implementation

Pattern 2: Audio Engine Integration

• Location: DadBrain/DadBrain/ViewModels/VoiceRecorder.swift
• Description: AVAudioEngine configuration for real-time audio buffer capture
• Frequency: 1 complete implementation

Pattern 3: Recognition Request Handling

• Location: DadBrain/DadBrain/ViewModels/VoiceRecorder.swift
• Description: SFSpeechAudioBufferRecognitionRequest with result processing
• Frequency: 1 complete implementation

TODO: Implementation

This skeleton skill needs Phase 2 refinement. Areas to develop:

Workflow Instructions

1. Add speech recognition permission to Info.plist
2. Request authorization from SFSpeechRecognizer
3. Configure AVAudioEngine for audio input
4. Create SFSpeechAudioBufferRecognitionRequest
5. Handle recognition results and errors

Best Practices

• When to start/stop recognition
• Handling partial vs final results
• Managing audio session categories
• Error handling for recognition failures

Common Pitfalls

• Missing microphone usage descriptions
• Not handling audio session interruptions
• Forgetting to stop recognition when done
• Ignoring availability checks (SFSpeechRecognizer.isAvailable)

Suggested Resources

Based on codebase analysis, consider adding:

Scripts (scripts/)

• speech-permission-handler.swift - Handles authorization request flow
• recognition-tester.swift - Test script for speech recognition setup

References (references/)

• speech-framework-guide.md - Apple's Speech Framework documentation
• audio-session-configuration.md - AVAudioSession setup patterns
• recognition-results-handling.md - Processing speech results

Assets (assets/)

• permission-templates/ - Info.plist usage descriptions for speech
• recognition-templates/ - Boilerplate recognition setup code

Evidence from Codebase

Refinement Priority

Score: 6.6/10 Priority: Medium

Refinement Tasks

1. Research Speech Framework documentation
2. Create speech-permission-handler script
3. Document audio session configuration
4. Add examples for different recognition scenarios
5. Create troubleshooting guide for common issues

---

Skeleton generated by ContextHarness /baseline Run Phase 2 skill refinement to complete implementation