Moshi AI vs SplitSong

Moshi AI is a speech-native conversational model from Kyutai, a Paris-based open-science research lab. Instead of chaining speech recognition, text generation, and text-to-speech, Moshi processes audio directly and holds full-duplex voice conversations with minimal latency.

Its multi-stream design runs separate channels for the user, Moshi's spoken output, and an Inner Monologue text stream that improves coherence. That setup lets Moshi listen and talk at the same time, handle overlaps, interruptions, and backchanneling like a real conversation rather than rigid speaker turns.

Moshi is built on Helium, a 7B language model, and Mimi, Kyutai's neural audio codec. Weights and inference code ship for PyTorch, Rust, and MLX, and you can try it in the browser at moshi-chat.kyutai.org. Researchers, voice AI developers, and anyone building real-time spoken interfaces will find the most value here.

SplitSong is a web tool that separates a song into individual instrument and vocal tracks using AI. Upload an audio file or paste a YouTube link, and the service returns downloadable stems such as drums, bass, instrumental, and vocals.

The homepage demonstrates the workflow with a Michael Jackson Thriller example, showing separate download links for each stem type. Users need to sign in before uploading their own tracks.

SplitSong was created by Mark Doppler and focuses on a simple stem-splitting workflow for musicians, remixers, and creators who want isolated tracks without manual editing.