(Tuesday September 13, 2023, New York) – Known for leading the frontier in speech translation, KUDO has made further improvements to user experience by enhancing its AI Speech Translator with adaptive voice technology. Weeks after upgrading the KUDO AI engine with the integration of LLM and generative AI, this release has notably augmented the naturalness and adaptability of its translated voices. The feature is available for all KUDO AI users.
The effectiveness of speech translation largely depends on the authenticity of its voice. Within the sphere of real-time speech translation—an area of expertise for KUDO—achieving genuine-sounding voices is paramount. KUDO AI’s most recent upgrade therefore includes the integration of advanced “adaptive voices” technology. The goal? To make the translated voice dynamically adapt to the pace of each individual speaker, and to the unique features of the way they speak. The result is a more natural, human, and dynamic user experience that ensures minimal possible latency—another unique feature of KUDO AI.
“Real-time, continuous translation requires the system to keep pace with the speaker and adapt the overall translation strategy consequently,” states Claudio Fantinuoli, CTO of KUDO. “At KUDO, we are all about the quality and personalization of the translation process for our clients. One element to achieve this is our newly introduced Adaptive Voices, but our R&D efforts certainly won’t stop there; with ongoing experiments to integrate emotions into interpretations, the ambition is to mirror the words and the sentiments behind them.”
Isabel Olmos, Linguistic Analyst at KUDO, underscores the importance of this endeavor. “What we are accomplishing with KUDO AI differs in complexity to voice dubbing. The kind of emotions we are eager to bring into our speech translation are more subtle and focus on the ability to speak with a voice that adapts intonation and other spoken language features to mirror the meaning of what has been said.”
Among its upcoming releases, KUDO is also planning a novel project focusing on profiling speakers’ voices and transplanting some unique characteristics from the original voice into the translated one in real time. This partnership initiative with leaders in the industry seeks to go beyond just gender matching—a feature already available in KUDO AI. The intended outcome is for users to experience the translated voice as if the speaker is conversing directly in their language. This kind of technology has emerged for offline content in the last few months, and the results are remarkable. Our challenge is to bring this in a real-time setting, with the necessity to profile the voices on the fly and render them immediately.
KUDO is the world leader in providing real-time multilingual solutions that enable people to communicate effortlessly in any language—on any platform. Their network of 12,000 professional language interpreters, combined with their ground-breaking Speech Translator, empower organizations of all sizes to collaborate more efficiently, with greater inclusivity, and on an international scale. KUDO Inc. is a New-York based technology start-up founded and managed by language and conferencing industry insiders seeking to create a world in which everyone has the power to understand and be understood in their own language. More info at www.kudoway.com.