Shypd.aiThe largest open index of AI toolsSpeech-to-speech is an open-source AI tool that builds local voice agents with open-source models. It provides a modular, cascaded pipeline for speech-to-speech conversion, supporting various STT, LM, and TTS models.
Trending
About
Speech-to-speech is an open-source project by Hugging Face designed to build local voice agents using a modular, cascaded pipeline approach. It integrates Voice Activity Detection (VAD), Speech-to-Text (STT), Language Model (LM), and Text-to-Speech (TTS) components. The tool emphasizes flexibility, allowing users to leverage various open-source models from the Hugging Face Hub, including Whisper, Parakeet TDT, MLX LM, MeloTTS, and ChatTTS. It supports multiple deployment methods, including server/client, WebSocket, and local execution, with optimized settings for Apple Silicon. The project also offers multi-language support, enabling single-language conversations or automatic language switching, making it a versatile platform for developers and researchers working with voice AI.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
