HONOR Introduces On-Device Speech Recognition and Real-Time Translation Model Supporting Arabic

Dubai - Emirates247: HONOR has launched a large speech language model that runs directly on-device as part of the international release of its new Magic V5 smartphone. The technology focuses on multilingual speech recognition and real-time translation without relying on cloud infrastructure or requiring an internet connection.

Features and Efficiency

According to the company, the model enables speech recognition and instant translation with low latency, while reducing memory consumption to around 800 MB, compared to the previous requirement of 3–4 GB. Six language packs—Arabic, Chinese, English, German, French, Spanish, and Italian—have been consolidated into a single package, removing the need to download each language separately.

HONOR says the on-device approach also addresses privacy concerns, particularly in phone calls, as most translation solutions today depend on cloud servers. Company data shows that the new system improves inference speed by 38% and translation accuracy by 16% compared to conventional methods.

Research Recognition

Two of HONOR’s research papers were accepted at INTERSPEECH 2025.

The first paper, titled “MFLA: Monotonic Finite Look-ahead Attention for Streaming Speech Recognition”, introduces an approach combining Continuous Integrate-and-Fire (CIF) prediction with a Wait-k strategy to enable low-latency and high-accuracy speech recognition on devices.

The second, in collaboration with Shanghai Jiao Tong University, presents a dual-scale “speculative sampling” model designed to accelerate multilingual speech translation while maintaining efficiency on resource-limited devices.

HONOR operates in the smart device sector, with a portfolio that includes smartphones, laptops, tablets, and wearables, all supported by artificial intelligence technologies.