ElevenLabs has introduced Speech to Speech (STS), a voice conversion tool that allows users to transform recordings to sound as if spoken by another character, with control over emotions, tone, and pronunciation, enhancing the capabilities of text-to-speech (TTS) systems. STS is particularly useful for extracting more emotions from premade voices and providing a reference for speech delivery, improving the precision of editing outputs. The company is also updating its premade voices, planning to add over 20 new voices and providing information on their availability. Additional updates include the introduction of Eleven Turbo v2 for real-time interactions, adherence to industry-standard audiobook submission guidelines, and the integration of a Pronunciation Dictionary, which supports IPA, CMU, and word substitutions. These enhancements aim to improve voice variety, customization, and the overall user experience on the platform.