
from qwen3_tts_rs219
Generate speech audio from text or clone voices from reference audio using Qwen3 TTS binaries and models; supports multiple named speakers, English and Chinese.
Provides binaries and workflow to synthesize speech from text and clone voices using Qwen3 TTS. Includes pre-packaged models, reference audio for named speakers, and clear command examples for TTS and voice cloning (ICL) with required parameters and output file locations.
Use this skill when a user asks to convert text to speech, generate audio files in a named voice, or perform voice cloning from a reference WAV and transcript. Also useful for batch TTS generation and local inference workflows where low-level control over model paths and environment variables is needed.
Inferred: agents that can run local binaries and manage files (shell-capable coding assistants), and audio toolchains for TTS/voice cloning tasks.
Qwen3 TTS skill for text-to-speech generation and voice cloning using local Qwen3 model binaries. SKILL.md is well-documented with clear parameter tables, examples for both TTS and voice clone modes, and platform-specific setup instructions. No bundled scripts were available for execution testing — the skill relies on pre-compiled binaries and model files hosted in the repo. Shell variable usage is properly quoted. No security concerns found.
No scripts to execute — skill depends on compiled binaries (tts, voice_clone) and model directories that weren't included in the fetch. SKILL.md quality is good with clear workflow, but the skill can't be validated without the actual binaries. Voice cloning reference audio for celebrity voices (Trump, Elon Musk) raises minor ethical considerations but not security issues.