Mimi
A low-cost real-time speech-to-speech model introduced by ElevenLabs, focused on natural conversation.
Timeline (last 90 days)
ElevenLabs launched the real-time speech-to-speech models Max and Mimi; Max scored 98.8% on Big Bench Audio and begins speaking in about 0.57 seconds.
Discrete token schemes (such as EnCodec, SoundStream, and Mimi) require predicting a set of residual vector quantization codebooks at each timestep, and in Mistral's scenarios having more than 30 codebooks significantly increases generation latency.
All Facts (2)
Discrete token schemes (such as EnCodec, SoundStream, and Mimi) require predicting a set of residual vector quantization codebooks at each timestep, and in Mistral's scenarios having more than 30 codebooks significantly increases generation latency.
50%UnverifiedElevenLabs launched the real-time speech-to-speech models Max and Mimi; Max scored 98.8% on Big Bench Audio and begins speaking in about 0.57 seconds.
50%