ElevenLabs: New AI Audio Watermark & Detector Launched

AI voices are getting harder to spot, but a new feature from ElevenLabs could change that. The company is integrating Google DeepMind’s SynthID watermarking technology into its AI-generated speech. Here's the thing: this watermark is embedded directly into the audio, making it inaudible to listeners. ElevenLabs says it survives common edits like trimming, compression, and file conversions. This means the watermark stays with the audio, even after it leaves ElevenLabs' servers. What's interesting is that ElevenLabs is also launching a free Audio Detector. Anyone can use it to check if a recording was created using their platform. The bottom line: as AI-generated audio becomes almost indistinguishable from human speech, this technology aims to help listeners determine if a clip is real or AI-generated.