From the source
ElevenLabs is launching Eleven Realtime Translation, a live voice translation model available in alpha for ElevenAPI enterprise customers, so each person speaks in their own language and hears the other in theirs while the conversation happens.
- The company says the model answers a scaling problem for large customer operations, where a caller in one language often waits for an agent who understands them.
- It runs by default on Eleven v4 Turbo, described as ElevenLabs' low-latency speech model optimized for realtime applications, with automatic language detection across more than 90 languages.
- Users pick any voice from a library of more than 17,500 or build an Instant or Professional Voice Clone, and that voice stays consistent for the whole session.
- Teams can tune how quickly translation starts, trading higher-quality translation for complex calls such as billing or claims against the lowest latency for faster back and forth.
- Deutsche Telekom is already building new calling features with ElevenLabs as part of its Magenta AI Call Assistant.
