Convert AI responses into spoken audio using the macOS say command.
npx skills add https://github.com/epologee/leclause-skills --skill saysay
Please help me install this Agent Skill. Skill: saysay Source: https://github.com/epologee/leclause-skills/tree/main/packages/saysay/skills/saysay Command: npx skills add https://github.com/epologee/leclause-skills --skill saysay
Replace screen-centric AI output with spoken output so users can receive responses hands-free, improve accessibility, and hear concise summaries instead of reading long or dense text.
Activate say mode by issuing the command /saysay so Claude speaks each response aloud using the system voice at a comfortable speaking rate.
High-intent search queries and answers about installing and using this skill.
Text-to-speech output for AI responses on macOS is enabled by activating say mode, which routes each response through the macOS say command at a default speaking rate of 240 for hands-free audio delivery.
You can correct mispronunciations during text-to-speech output by using a per-user phonetic dictionary, stored as a JSON file in your XDG_DATA_HOME directory, to define exact pronunciations for proper names and technical terms.
Backgrounded speech execution prevents text-to-speech calls from blocking the interactive CLI workflow, allowing multiple parallel sessions to run simultaneously with distinct context labels distinguishing each audio output.
Hearing code review summaries aloud uses macOS text-to-speech to narrate key findings from test suites or code reviews, keeping full textual details on screen for later inspection while replacing screen-centric reading with spoken output.
The macOS say command is required to enable speech output for accessibility workflows, serving as the underlying system voice engine that serializes AI responses into spoken audio at a comfortable rate.
seedprodConvert text to speech via ElevenLabs with a macOS-style command-line interface.
bishwashereTranscribe audio to text with Whisper and synthesize speech with 11Labs.
auxi-wardrobeOrchestrate multiple Codex sessions for parallel research, implementation, review, and debug workflows.
mikeyobrienConverts text to speech on macOS using the built-in utility.
