Can I control the emotion of the AI voice?

Yes, when using the eleven_v3 model, you can use tags like [excited], [whispers], or [mischievously] to guide the emotional delivery of the text.

How do I handle mispronounced words in Sag?

You can use phonetic respelling (e.g., 'key-note' for keynote) or adjust the --normalize flag to help the model interpret units, URLs, and complex names correctly.

Does Sag support languages other than English?

Yes, Sag supports the eleven_multilingual_v2 model and includes a --lang flag to guide normalization for various languages like German, French, and more.

What is required to use the Sag skill?

You must have the 'sag' CLI tool installed (typically via Homebrew) and a valid ELEVENLABS_API_KEY set in your environment variables.

Sag ElevenLabs TTS

Name: Sag ElevenLabs TTS
Author: HarleyCoops

byHarleyCoops

内容管理

Generates high-quality, expressive text-to-speech audio using ElevenLabs with a simple macOS-style command-line interface.

关于

Sag is a specialized Claude Code skill that integrates the 'sag' CLI to bring ElevenLabs' powerful text-to-speech capabilities directly into your AI workflow. It allows Claude to generate audio files using advanced models like eleven_v3, multilingual_v2, and flash_v2_5. The skill provides granular control over vocal delivery through expressive tags such as [whispers], [shouts], or [sarcastic], and features sophisticated pronunciation handling via phonetic respelling and language-specific normalization. It is ideal for developers creating voice-enabled applications, content creators needing quick high-fidelity voiceovers, or users who want Claude to respond with specific vocal personas.

主要功能

Support for custom voice IDs and specific language biases for normalization
Access ElevenLabs TTS models including v3 Expressive and Flash v2.5
0 GitHub stars
Manage pronunciation with phonetic respelling and normalization rules
Control vocal delivery with tags like [whispers], [shouts], [laughs], and [sings]
Generate and output MP3 files directly within the Claude chat interface

使用场景

Generating realistic vocal personas for AI-driven characters or virtual assistants
Prototyping and testing audio scripts with precise delivery and timing control
Automating the creation of high-quality voiceovers for tutorials and marketing content

关于

主要功能

Support for custom voice IDs and specific language biases for normalization
Access ElevenLabs TTS models including v3 Expressive and Flash v2.5
0 GitHub stars
Manage pronunciation with phonetic respelling and normalization rules
Control vocal delivery with tags like [whispers], [shouts], [laughs], and [sings]
Generate and output MP3 files directly within the Claude chat interface

使用场景

Generating realistic vocal personas for AI-driven characters or virtual assistants
Prototyping and testing audio scripts with precise delivery and timing control
Automating the creation of high-quality voiceovers for tutorials and marketing content