To achieve this level of efficiency, Ateme integrated a new processing step directly connected to Google Cloud's managed services into its software platform. Once video files are transcoded, the system automatically triggers a call to Vertex AI and the Gemini models to perform audio transcription, precise timecode spotting, and subtitle generation in the desired language. These are then reformatted into the industry-standard SRT format through an internally developed script. Fully integrated into the existing workflow, this additional feature seamlessly fits into clients' interfaces, with no technical complexity or disruption to their usual processes.

Beyond the immediate operational gains, this transformation also opens the door to unprecedented opportunities, as Jules Mermet-Husson highlights: "It becomes possible to massively enrich existing VOD catalogs, accelerate content availability, broaden linguistic coverage, and also meet regulatory accessibility requirements, especially for the hearing impaired. Market feedback has been very positive: at NAB Show in Las Vegas, we saw a strong interest in this solution, particularly from international broadcasters seeking scalability."

In fact, the transformation carried out by Ateme generates direct benefits for its clients: better profitability of existing catalogs through the ability to subtitle massively at a lower cost; easier access to new linguistic markets; increased and more cost-effective compliance with regulatory accessibility requirements; and, ultimately, an enriched user experience for wider and more diverse audiences.

On the underlying infrastructure side, the project relies on a robust foundation. Ateme's file transcoding application, Titan File, is containerized as microservices and deployed through Google Kubernetes Engine. Media files are stored in Google Cloud Storage, from which they are processed by Vertex AI. Thanks to this distributed, flexible, and scalable infrastructure, Ateme did not have to rewrite its existing code to adapt; only an integration layer was needed. "The transition was smooth," says Jules Mermet-Husson.

This subtitling project is the first building block of Ateme's stated ambition to bring increasingly innovative services to its clients. The company is already working on future use cases leveraging the same technological foundation. Audio description, for example, is in their sights: it requires visual and contextual analysis capabilities that the most recent multimodal models are beginning to offer. In the longer term, Ateme plans to leverage this momentum to enrich its offerings around content personalization, smart clipping, and advanced recommendations.

By basing their partnership on continuous innovation, Ateme and Google Cloud affirm their position as pioneers in the convergence of video and artificial intelligence. "Faced with the explosion of content volumes, it's no longer enough to have good technology: you have to be able to innovate quickly, at scale, and with the right partners. Google Cloud's generative AI gives us the means to automate what was once a craft. Together, we are building an agile, open, and scalable platform that transforms every technical constraint into an opportunity for value creation," concludes Jules Mermet-Husson.