What is the Sag skill for Claude Code?

Sag is a specialized skill that allows Claude to use the ElevenLabs API to generate high-quality speech with a simple command-line interface similar to the macOS 'say' command.

Does Sag support emotional voice tags?

Yes, when using the eleven_v3 model, you can use tags like [whispers], [shouts], [laughs], and [sarcastic] to guide the delivery of the speech.

Which ElevenLabs models are available in Sag?

Sag supports eleven_v3 for expressiveness, eleven_multilingual_v2 for stability, and eleven_flash_v2_5 for low-latency tasks.

How do I authenticate ElevenLabs with Sag?

You need an ElevenLabs API key, which can be provided via the ELEVENLABS_API_KEY or SAG_API_KEY environment variables.

Can I save the generated audio to a file?

Yes, you can use the -o flag followed by a file path (e.g., sag -o output.mp3 'text') to save the generated audio instead of just playing it.

Sag ElevenLabs Voice Synthesis

Name: Sag ElevenLabs Voice Synthesis
Author: MillionthOdin16

byMillionthOdin16

Gestión de Contenido

Generates high-fidelity speech from text using ElevenLabs models with a simple, Mac-style command-line interface.

Acerca de

The Sag skill integrates powerful text-to-speech capabilities into Claude Code, allowing the AI to generate and play audio using ElevenLabs' advanced voice models. It provides comprehensive control over the vocal output, including support for expressive v3 audio tags like whispering, shouting, and laughing. This skill is ideal for developers building voice-enabled applications, creators needing rapid narration prototyping, or users who want Claude to provide character-driven audible responses directly from the terminal.

Características Principales

0 GitHub stars
Access to multiple models including eleven_v3, multilingual_v2, and flash_v2_5
Support for expressive v3 audio tags including [whispers], [shouts], and [sings]
High-fidelity text-to-speech via ElevenLabs API integration
Local audio playback and direct export to MP3 files
Advanced pronunciation control with respelling and normalization rules

Casos de Uso

Prototyping voiceover content with specific emotive pacing and delivery
Testing ElevenLabs voice configurations and pronunciation fixes in real-time
Generating character-based audio responses for interactive CLI applications

Acerca de

Características Principales

0 GitHub stars
Access to multiple models including eleven_v3, multilingual_v2, and flash_v2_5
Support for expressive v3 audio tags including [whispers], [shouts], and [sings]
High-fidelity text-to-speech via ElevenLabs API integration
Local audio playback and direct export to MP3 files
Advanced pronunciation control with respelling and normalization rules

Casos de Uso

Prototyping voiceover content with specific emotive pacing and delivery
Testing ElevenLabs voice configurations and pronunciation fixes in real-time
Generating character-based audio responses for interactive CLI applications