Generates speech with voice cloning and emotion-driven visual-timing data via Inworld API endpoints.
npx skills add https://github.com/itechmeat/llm-code --skill inworld
Please help me install this Agent Skill. Skill: inworld Source: https://github.com/itechmeat/llm-code/tree/main/skills/inworld Command: npx skills add https://github.com/itechmeat/llm-code --skill inworld
This skill enables AI agents to generate natural-sounding speech using Inworld's TTS API, including voice cloning, emotion control, and timing data for lip-sync.
Use the inworld skill to synthesize speech from text with a cloned voice and request timing data for lip-sync.
High-intent search queries and answers about installing and using this skill.
Voice cloning and lip-sync data are generated by sending a POST /voice request to the Inworld TTS API with your text, voiceId, and timestampType, returning audioContent and timestampInfo. This process enables expressive speech synthesis with accurate viseme timing for avatar animation.
The Inworld TTS API returns timestampInfo containing word, phoneme, and viseme timing data when you set the timestampType parameter. This timestamped lip-sync data provides accurate viseme timing required for character and avatar animation workflows.
Emotion and delivery style can be controlled in text-to-speech synthesis by applying emotion markers and delivery styles to your Inworld API request. This produces expressive speech with visemes, allowing applications and agents to generate natural-sounding speech with specific emotional tones.
Text-to-speech audio can be streamed using the optional POST /voice:stream endpoint provided by the Inworld TTS API. You configure parameters such as voiceId and modelId to receive streaming audioContent, offering straightforward API-first integration for apps, bots, and games.
Cloning a voice from a short sample requires configuring parameters such as voiceId and modelId within the Inworld TTS API request. The API synthesizes expressive speech from your text input using the cloned voice profile, applicable to assistants and game characters.
The best way to integrate expressive speech synthesis into an AI agent is through the API-first Inworld TTS endpoints, using POST /voice with voiceId and modelId parameters. This returns audioContent and timestamped viseme data, enabling straightforward integration for bots and assistants.
tony2015116Synthesize and clone voices with MiniMax Voice API and FFmpeg.
MCKRUZAutomate character voice creation and lip-sync for video production.
GroundMountCompanyGenerate AI talking head videos with lip-sync using multiple model backends.
ScientiaCapitalFine-tune Llama, Mistral, Phi, or Gemma models with LoRA and 4-bit quantization.
