Vodafone and Ericsson trial network-level AI voice translation for mobile calls

This trial signals a shift in how mobile operators deliver value-added services by moving intelligence from the device to the core infrastructure. By utilizing network-level AI voice translation, carriers can offer premium features to all subscribers regardless of their hardware, potentially creating a new revenue stream or a tool for customer retention. For the broader streaming and communications ecosystem, this demonstrates that low-latency media processing is increasingly viable at the network edge. As these capabilities move out of the lab, the industry should watch for Vodafone to announce a commercial rollout timeline or specific pricing tiers for these AI-enhanced calling features.

Ericsson has been building out its AI integration with the IP multimedia subsystem as part of a broader strategy to position network infrastructure as a platform for media services. In May 2026, Bitmovin's annual Video Developer Report found that 98 per cent of video professionals now use AI or ML in their workflows, with audio transcription, translation, and foreign dubbing cited as the most common applications at 48 per cent of respondents. That demand signal from the video side underscores why telcos like Vodafone see network-level translation as a commercially viable service rather than a lab curiosity.

The patent and licensing landscape around AI-enhanced media processing has shifted materially in the past 18 months. Access Advance launched the Video Distribution Patent Pool in January 2025, and Nokia and Amazon settled their global streaming video patent litigation in March 2025, reducing uncertainty for companies deploying codec and media-processing technologies at scale. For Ericsson, which holds substantial patent portfolios in both telecoms and media processing, these settlements clarify the commercial environment in which network-level AI services can be monetized without triggering cross-licensing disputes.

On the competitive side, the race to embed AI into network media planes is not limited to Ericsson and Vodafone. Mux launched its Robots product in 2026, offering first-party AI moderation and analysis that runs natively alongside video assets inside its platform, demonstrating the same architectural principle of moving intelligence closer to the media rather than requiring separate orchestration layers. Meanwhile, Bitmovin announced in May 2026 that MUBI selected its VOD Encoder to replace a legacy on-premises encoding stack, supporting multi-codec strategies including AV1, showing that cloud-native media processing platforms are consolidating AI capabilities directly into encoding pipelines. Both moves mirror the Vodafone-Ericsson approach of embedding processing natively rather than bolting it on at the application layer, though they operate in the OTT and developer-tools segment rather than the telco core network. The industry is also seeing ITU and Huawei define AI-native network specifications to standardize these deployments.