Introducing KIARA Voice Studio 🎙️ — AI Audio & Video Voice Conversion

An individual developer released KIARA Voice Studio, an AI audio and video voice-conversion application built on Seed-VC v2, PyTorch, BigVGAN, FFmpeg and Gradio, running cloud inference on Hugging Face ZeroGPU. The tool converts source audio toward a reference speaker's voice and supports voice conversion in uploaded videos through an interactive Gradio interface. The developer, who named the app after his daughter, said he is also starting KIARA-VC Mini, a personal research effort to train a smaller voice-conversion model, and asked the Hugging Face community for feedback on conversion quality, audio-video synchronization and usability while urging users to process only voices they own or have permission to use.

Hello Hugging Face Community I’m an individual developer exploring Artificial Intelligence, Machine Learning, and Deep Learning. I’d like to share KIARA Voice Studio , my AI-powered audio and video voice-conversion application. KIARA is named after my daughter, and building this application has been an important part of my AI learning journey. Converts source audio toward a reference speaker’s voice Supports voice conversion in uploaded videos Uses Hugging Face ZeroGPU for cloud-based AI inference Provides an interactive Gradio interface Seed-VC v2 PyTorch BigVGAN FFmpeg Gradio Hugging Face ZeroGPU I’m also starting KIARA-VC Mini , a personal learning and research journey to understand speech AI and eventually train my own smaller voice-conversion model. I’d appreciate feedback from the Hugging Face community, especially on: Voice-conversion quality Audio-video synchronization Application usability Suggestions for improving speech conversion Please use only voices you own or have permission to process. Thanks for taking a look VoiceConversion SpeechAI DeepLearning HuggingFace SeedVC