xfyun-voiceclone

What problem does it solve?

This Skill automates the complex and error-prone process of creating a custom text‑to‑speech voice from example recordings and then synthesizing speech in that cloned voice, removing the need to manually craft auth, multipart uploads, and WebSocket TTS plumbing.

Core Features & Use Cases

• End-to-end training workflow: obtain training text, create a training task, upload audio (file or URL), submit the task, and poll for completion to get a voice resource ID.
• Synthesis with cloned voice: connect to iFlytek TTS over WebSocket, synthesize text into mp3/pcm/speex/opus outputs, and adjust speed, volume, pitch and sample rate.
• Use Cases: create personalized voice assistants, generate demo audio or localized narration from a speaker's recordings, and prototype voice UX without external dependencies.

Quick Start

Use the xfyun-voiceclone scripts to train a voice from a local WAV file and synthesize a short test phrase to an MP3 using your XFYUN_APP_ID, XFYUN_API_KEY, and XFYUN_API_SECRET.