ifly-voiceclone-tts

What problem does it solve?

Clone a custom voice from audio samples and synthesize speech with the cloned model, enabling personalized voice experiences without relying on generic TTS voices.

Core Features & Use Cases

• Two-phase workflow: Train a voice model by uploading audio samples, then synthesize speech using the resulting res_id.
• Self-contained tooling: Pure Python stdlib-based implementation with no pip dependencies; environment variables IFLY_APP_ID, IFLY_API_KEY, and IFLY_API_SECRET control access.
• Real-world scenarios: Create branded IVR prompts, narration for videos, or accessibility features with your own voice.

Quick Start

Run the training workflow to create a task, upload audio, submit training, and obtain a res_id, then use it to synthesize speech with your cloned voice.