Speech Swift
Speech Swift is a AI agent skill for audio, voice, and music, published by soniqo.
Complete speech recognition, synthesis, and understanding stack running locally on Apple Silicon via MLX Swift and CoreML.
Is it any good?
Effectiveness · benchmark pending
We haven't benchmarked Speech Swift yet, so its score is provisional — computed from safety, maintenance, and documentation only, with popularity given no weight. When we benchmark it, we measure whether it produces materially better output than a frontier model alone, and that result leads the score. How we measure.
About Speech Swift
Description · AI-summarised from the soniqo README
Speech Swift is a production-grade Swift package providing on-device speech AI models for macOS and iOS. It includes speech-to-text (Qwen3-ASR, Parakeet, Omnilingual ASR across 1,672 languages), text-to-speech (Qwen3-TTS, Kokoro, CosyVoice, VibeVoice), voice activity detection, speaker diarization, speech enhancement, music generation, and an on-device LLM chat. All models run locally on Apple Silicon (M1/M2/M3/M4) with no cloud dependencies, API keys, or data leaving the device. Supports both MLX and CoreML backends, includes SwiftUI components, CLI tools, and demo applications.
Where it fits
Browse related skills
Install Speech Swift
Install
Install via agent tool
# install via agent tool
# see vendor docsInstall with your AI
Zero install · paste into ChatGPT, Claude, Cursor…
Install the "Speech Swift" skill into my project by following all instructions at https://skillsdirectory.co/s/soniqo-speech-swift
Your assistant reads /s/soniqo-speech-swift and follows it — fetching the files from the source, placing them, and confirming before it runs anything. No account needed. How this works.
Frequently asked
FAQ
What does Speech Swift do?
Complete speech recognition, synthesis, and understanding stack running locally on Apple Silicon via MLX Swift and CoreML.
How do I install Speech Swift?
Speech Swift is a AI agent skill. Primary install path: # install via agent tool # see vendor docs
What is the Skill Score for Speech Swift?
Speech Swift has a Skill Score of 6.4/10 (Solid). It is currently a provisional score — Speech Swift hasn't been benchmarked yet, so it weights safety 75/100 (40%), maintenance 50/100 (30%), and documentation 71/100 (30%). Popularity carries zero weight until effectiveness is measured. See methodology.
Is Speech Swift free?
Speech Swift is published under the Apache-2.0 license. See the vendor's source repository for any usage limits or paid tiers.
Related skills in Audio, voice, and music
Related
Claude AI Music Skills
A Claude Code plugin that orchestrates music album production across research, lyric writing, Suno prompts, mastering, and release prep with multi-model routing and quality gates.
Impeccable
Frontend design skill for Claude and other AI agents with 23 commands, 7 domain references, and anti-pattern detection to enforce consistent, high-quality design.
MeiGen AI Design MCP
MCP server that enables AI agents to generate professional images and videos using 9 leading models, with parallel batch orchestration and 1,446 curated prompts.
Embed this score
For maintainers · always-current, links back
Put it in your README — it updates as the score changes.[](https://skillsdirectory.co/skills/soniqo-speech-swift)
<a href="https://skillsdirectory.co/skills/soniqo-speech-swift"><img src="https://skillsdirectory.co/badge/soniqo-speech-swift.svg" alt="Scored by skillsdirectory" height="28" /></a>