A Human-centered Approach to AI Dubbing
Audiovisual Localization SIG Meeting - May 2026
The field of AI dubbing has sparked curiosity and various application approaches within the industry following the significant advancements in speech synthesis quality. However, as with other speech technologies in audiovisual translation, a human-centered approach is needed to refine the unprocessed output of generic AI systems. In other words, trained and experienced experts are required to enhance artificially generated voices, making them as authentic as possible in the target language while conveying the essence of the original material. This is the only way to guarantee that the quality meets the expectations of the intended customer segment.
The growing international demand for AI Speech Editors requires a profile that encompasses audiovisual post-production skills and goes beyond. As well as proficiency in producing text for audio description, voice-overs and dubbing, the role demands the technological and vocal expertise required to unleash the maximum potential of text-to-speech technology.
Starting with textual adaptation for high-quality studio and sound production recordings, new competencies range from appropriately casting artificial voices for target language versions, to professionally applying the required software environments, and understanding the potential and limitations of artificial voices. Unlike more generic AI outputs, these skills ensure that the random voice output of 'black box' AI applications can be controlled to achieve the best possible quality for the desired result.
This presentation reflects on the current status and latest advancements in AI voice production for media localization, as well as its limitations, from the perspective of audiovisual translators working with synthetic speech. It serves as a starting point for discussion about the novel and expanded scope of tasks and skills relating to the professionalization of AI Speech Editors. Finally, it aims to shed light on the expectations associated with this increasingly in-demand role.
Key Takeaways:
- AI dubbing is not a tool — it’s a workflow that needs specialists.
- This is trainable, scalable, and can be professionalized.
- It is vital to stipulate the level of expectation for the quality of AI voice productions.
Please join us on Wednesday, 27 May for this meeting of GALA's Audiovisual Localization SIG, featuring guest speakers Alexander Kurch and Yasmine Bendimered of AdaptGlobal.io.
Registration Options
|
Registration Options
|
Price |
|---|---|
|
MEMBER TICKET
|
FREE |
|
NON-MEMBER TICKET
|
FREE |
SIG Meeting Speakers
Alexander Kurch, German Language Advisor, Adapt
Alexander Kurch has worked in audiovisual translation and media accessibility since 2012, specializing in speech-to-text interpreting, subtitling, audio description, voice-over, and dubbing. Alongside his practice, he is an independent researcher, author, lecturer, and trainer in audiovisual translation and interpreting. His focus lies in combining accessible communication with speech and language technologies, exploring speech recognition, machine translation, speech synthesis, and AI voice optimization. More recently, he has expanded into AI tool testing, generative AI, prompt engineering, modular tech stacks, and AI-driven workflow automation in media localization.
Yasmine Bendimered, French Language Lead, Adapt
Yasmine is French Language Lead at Adapt. She is a subtitle translator and QC specialist with experience on projects for Amazon, Netflix, and Paramount. She ensures high-quality, culturally adapted subtitles.
GALA Special Interest Groups
GALA SIGs are global communities that connect members on topics of shared interest led by volunteer moderators. Meet the moderators for the Audiovisual Localization SIG.
Michael Anderson, Welocalize
Michael helps global organizations deliver multimedia and eLearning content that resonates across languages, cultures, and platforms. Over the past 15+ years, I’ve led strategy and execution for localization workflows, multimedia production, and AI-enabled solutions that scale from pilot to enterprise adoption.
Andrea Ballista, VOISEED SRL
Andrea has been working in the game industry for more than 25 years. Starting from the passion of music, songwriting and singing, he holds a Computer Music graduation. After having co-founded Binari Sonori in 1994, he has been Audio Director for Keywords Studios from 2015 until the end of 2019. In 2020, he has co-founded Voiseed to proceed into research and development of deep learning based emotional virtual voices.
Liza Dziahel, Alconost
Liza is a Marketing Manager at Alconost, a global translation and localization agency with over 600 native-speaking translators working in more than 100 languages. With a passion for thought leadership content and all things global, she is responsible for managing the company's digital communication channels, including the blog, social media, and community discussions.
Carlo Eugeni, University of Leeds
I graduated in Conference Interpreting and Translation at the University of Bologna in 2003. In 2009 I got my Ph.D. in English for Special Purposes with a thesis on live subtitling through respeaking. I am associate professor of audiovisual translation at the University of Leeds (UK), where I teach conference interpreting and respeaking and subtitling for the Deaf and the Hard-of-Hearing.
I founded the biennial International Symposium on Live Subtitling and Accessibiity, and am the chairman of the Intersteno scientific committee. In 2016, I co-founded the international journal on linguistic and cultural mediation CoMe and in 2020 the journal of professional reporting and transcription Tiro. In 2018, I received the Italian association of cochlear implant users APIC honorary award for my commitment in promoting the empowerment of deaf people.
As a professional I have worked as a freelance simultaneous interpreter, a technical translator, a parliamentary reporter, and a professional intralingual and interlingual respeaker. I have published extensively in the fields of Media Access and Audiovisual Translation, and am currently a researcher of the Horizon-UKRI project on easy language iDEM.
Join GALA SIGs
Our special interest groups are global communities that connect members on topics of shared interest. Explore and join GALA SIGs.
For More Information:
