What problem does it solve? Changing the voice in an existing audio recording normally requires re-recording with voice talent, which is slow and costly. This Skill converts any speech recording into a different voice while preserving the original speech content and emotion, using ElevenLabs speech-to-speech models via the inference.sh CLI. ## Core Features & Use Cases - Speech-to-Speech Conversion: Transform any audio recording into one of 22+ premium voices (e.g., george, aria, brian) while keeping the original words and emotional delivery. - Multilingual and English Models: Choose between eleven_multilingual_sts_v2 supporting 70+ languages or eleven_english_sts_v2 optimized for English speech. - Configurable Output: Select output audio formats such as mp3_44100_192 for downstream use in video or podcast production. - Use Case: Record yourself reading a script on any microphone, then transform it into a professional British narrator voice and merge it into a video using the media-merger app. ## Quick Start Ask the AI to transform your audio file into a different voice, for example: change the voice in my recording at https://my-recording.mp3 to the aria voice using the ElevenLabs voice changer.
