October 2022 Summaries

Launching in November, Eleven Labs introduces a pioneering speech synthesis platform designed to generate high-quality, emotionally engaging long-form content for publishers and creators. This innovative tool distinguishes itself by offering lifelike, emotive synthetic voices that enhance storytelling for various text formats, including news articles, books, and academic papers. Unlike traditional algorithms that produce speech in a robotic manner, Eleven Labs’ model incorporates contextual understanding, allowing for natural pacing and tone across entire narratives. The platform, which is set to support multiple languages in the future, provides users with the option to use pre-defined synthetic voices or clone voices from samples. It aims to expand accessibility, boost productivity, and transform written content into audio experiences, with a beta-testing opportunity available through their website.

Eleven Labs recently participated in the INTERSPEECH 2022 conference, where they had the opportunity to present their advancements in voice cloning, speech synthesis, and voice conversion technologies. Engaging with both startups like Supertone and LOVO, as well as established companies such as Meta and Google, Eleven Labs showcased their progress toward developing an automatic dubbing tool slated for release next year. The company emphasized the importance of accurately preserving voice similarity and enhancing prosody and tonality to create natural-sounding synthetic speech. Their efforts were well-received, reinforcing their commitment to refining their text-to-speech tools. The conference provided a motivational boost and set the stage for further advancements in their technology and presentation methods, with aspirations to introduce production-quality dubbing tools that enable speakers to communicate in multiple languages.