Qwen3/Free Text-to-Speech and Voice Cloning

🆓 完全免费的本地文字转语音(TTS)与声音克隆技能。基于 Qwen3-TTS-1.7B 模型，支持 Apple Silicon，无需联网，保护隐私。可用于有声书制作、AI 角色配音、无障碍应用等场景。

Key things to know before installing: - The skill's documentation loudly claims 'no internet needed', but the demo and install actually download Python packages and a ~3GB model at runtime. Expect network activity and large downloads unless you pre-provide the model locally. - The install script runs the Homebrew install bootstrap (curl | bash) and then brew/pip installs. Review install_dependencies.sh before running and consider running it in a controlled environment (VM/container) if you are unsure. - Inspect the mlx-audio package and where load_model retrieves models from (model hub/URL) — verify license and trustworthiness of the model artifact before use. - If you need true offline operation, ask the skill author for explicit instructions to pre-download and place the model locally (and how to point load_model to local files) instead of allowing automatic downloads. - Make sure you have ~3GB free disk and sufficient RAM; and prefer manual installation steps so you can audit each command. Overall: the package is plausible for its stated function, but the mismatch about 'no internet' and automatic large downloads is a meaningful incoherence — treat this as suspicious and audit the remote sources before running.

Type: OpenClaw Skill Name: qwen3-free-tts Version: 1.0.0 The skill bundle provides a legitimate local Text-to-Speech and voice cloning tool for macOS (Apple Silicon) using the mlx-audio library and a Qwen-based model. The installation scripts (install_dependencies.sh, package.json) and the Python demo (voice_cloning_demo.py) perform standard setup and execution tasks consistent with the stated purpose, such as installing dependencies via Homebrew and pip. The documentation in SKILL.md includes comprehensive usage guides and ethical legal disclaimers, with no evidence of malicious intent, data exfiltration, or prompt injection attacks.