Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.
Repository Index
132,241 visible repositories in the index
1–9 of 9 · Most stars
Towards Open-source GPT-4o with Vision, Speech and Duplex Capabilities。
first base model for full-duplex conversational audio
Controllable Transcription. Verbatim ( every, filler, pause, stutter, vocal sound) , or intended ( what the speaker meant to say, optimized for readability) with word-level timestamps.
- asr
- audio
- detection
- filler
Real-time full-duplex speech recognition server, based on the Kaldi toolkit and the GStreamer framwork.
Haru: a low-latency, multilingual, full-duplex spoken dialogue model
Turn detection for full-duplex dialogue communication
- large-language-models
- natural-language-understanding
- real-time
- turn-taking
Open full-duplex voice models that hear your tone and speak with one. One GPU with vLLM-Omni, or fully offline on a Mac with MLX. Zero-shot voice cloning, OpenAI Realtime / GPT-Live compatible, free hosted API trial. Apache-2.0.
- full-duplex
- open-source
- paralinguistics
- realtime
Duo: A Reciprocal World Model for Full-Duplex Interaction