Voicebox is an open-source application designed for macOS that enables users to create voice profiles and synthesize audio locally. Utilizing the Qwen3-TTS model, this tool allows users to generate realistic voice audio directly on their device without the need for cloud services or subscriptions. Users can create voice profiles from short audio samples, enabling them to produce new audio outputs while maintaining the original tone and accent. The integration of Whisper enhances the experience by converting audio to text, making dialogue editing seamless and efficient.

In addition to voice synthesis, Voicebox features a local voice server through its REST API, allowing for easy automation of narration and notifications. The application supports system audio capture, enabling users to record and utilize sounds from various sources. Voicebox also includes a story generation feature, where users can create narrated content based on specified scenarios. While the cloning process is straightforward, there are areas for improvement, such as increased voice customization options and the ability to fine-tune audio samples for more realistic outputs.