skills.rest
Roy-wonji
Roy-wonjiCOMMUNITY
Explore all skills by this author →
GitHub

speech-recognition

Transcribe live microphone and local file speech using Apple's Speech framework.

Updated May 6, 2026
Search Tags:#authorization#speech recognition#transcription#SFSpeechRecognizer#AVAudioEngine#SpeechAnalyzer#on-device recognition
One-click install
npx skills add https://github.com/Roy-wonji/claude-config --skill speech-recognition-roy-wonji
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: speech-recognition
Source: https://github.com/Roy-wonji/claude-config/tree/main/skills/swift-ios-skills/skills/speech-recognition
Command: npx skills add https://github.com/Roy-wonji/claude-config --skill speech-recognition-roy-wonji

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Speech-to-text integration is hard because you must handle microphone/speech authorization, wire audio capture or file input, and correctly manage partial vs final transcription results.

Core Features & Use Cases

  • Live microphone transcription: Connect AVAudioEngine to SFSpeechAudioBufferRecognitionRequest and stream recognized text while recording.
  • Pre-recorded audio transcription: Recognize speech from local audio files using SFSpeechURLRecognitionRequest (server-based behavior).
  • iOS 26+ modern pipeline: Use the actor-based SpeechAnalyzer with modular SpeechTranscriber components, including required asset installation via AssetInventory.
  • On-device vs server recognition: Select offline/on-device recognition when supported and configure request options appropriately.
  • Practical result handling: Process best transcription, alternatives, confidence segments, punctuation, and contextual strings.

Quick Start

Ask the AI to provide an end-to-end Swift example that requests both speech and microphone permissions, starts live transcription from the microphone, and prints the final recognized text.

Frequently Asked Questions about speech-recognition

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I implement live speech-to-text transcription from a microphone in Swift?▼

You handle live speech-to-text by connecting AVAudioEngine to SFSpeechAudioBufferRecognitionRequest to stream microphone audio buffers to the recognizer. This Skill provides the logic to wire audio capture and process partial or final recognized text continuously.

What's the best way to transcribe a local audio file using Apple's Speech framework?▼

To transcribe local audio files, use SFSpeechURLRecognitionRequest for server-based speech recognition. This Skill orchestrates the request setup and processes the returned transcription results from pre-recorded audio files.

Does iOS 26 SpeechAnalyzer support on-device speech recognition?▼

Yes, iOS 26+ SpeechAnalyzer supports on-device speech recognition using an actor-based pipeline with modular SpeechTranscriber components. This Skill helps configure SpeechAnalyzer, manage required AssetInventory installations, and process AsyncSequence-based results.

How do I request microphone and speech recognition authorization for transcription?▼

You request microphone and speech recognition authorization by managing both permission flows before starting AVAudioEngine and SFSpeechRecognizer. This Skill provides the setup logic to safely handle authorization requests and the audio engine lifecycle for transcription.

When should I choose on-device recognition over server-based transcription?▼

Choose on-device recognition when offline support or lower latency is required, configuring request options appropriately. Server-based recognition via SFSpeechURLRecognitionRequest is typically used for pre-recorded files when on-device recognition is unavailable.

Related Skills

Explore More Software Engineering Skills→
FelixRauchFelixRauch

speech-recognition

Transcribe live microphone input and audio files via iOS Speech APIs.

Community
Intermediate
garochee33garochee33
3

transcribe

Transcribe audio and video recordings into text with speaker diarization.

Community
Intermediate
herdiansahherdiansah
23

social-listening-analyst

Analyze social media data to monitor brand mentions and assess sentiment.

Community
Intermediate
nuno7lopesnuno7lopes

sherpa-onnx-tts

Convert text to speech offline using sherpa-onnx with various voice models.

Community
Basic