Speech Translation Benchmark
Comparing real-time speech translation systems across translation quality, intelligibility, naturalness, latency, and speaker similarity.
Relay-1 model for speech-to-speech translation
Relay-1 is the model powering Pinch's real-time speech translation API, across 50+ languages, over a single WebSocket connection.
ElevenLabs vs Pinch for video dubbing
We dubbed the same video with both Pinch and ElevenLabs. Here's how they compare on translation quality, voice preservation, and pricing.
JEPA-v0: a self-supervised audio encoder for real-time speech translation
To build speech-to-speech translation that preserves voice, emotion, and timing, you need an encoder that captures more than just words. Here's how we're building one using JEPA.
Falcon v1.3 is live!
We shipped Falcon v1.3 with streaming voice cloning and faster time-to-first-translation for smoother live conversations.
Pinch Desktop: real-time translation that works with every meeting app
A macOS app powered by Falcon v1.3 that translates across 50+ languages in any meeting app - no plugins, no integrations.
Introducing Pinch Research
We are building the future of real-time speech translation.