WeChat Input Method has reached a milestone upgrade. Today, its iOS version officially released the major version update 3.0.0—directly upgrading the version number from 2.2.6 to 3.0, marking the product's full shift toward a next-generation voice-first interaction experience. The new version focuses on a complete reconstruction of voice input capabilities, not only upgrading the underlying large model to achieve "faster and more accurate" recognition effects, but also making significant progress in practicality, language coverage, and privacy protection.

The most notable feature of this update is deep support for 15 Chinese dialects, including Cantonese, Sichuan dialect, Henan dialect, Shaanxi dialect, Shanghai dialect, Suzhou dialect, Wuxi dialect, Nanchang dialect, covering mainstream accents across northern and southern regions. Users do not need to manually switch language modes; the system can automatically recognize and accurately transcribe, truly achieving a natural input experience of "speaking in dialect, outputting text."

For long-duration voice scenarios such as meeting notes, classroom lectures, and interviews, the 3.0 version completely removes the voice input time limit, supporting continuous transcription for several hours without interruption, meeting professional-level recording needs. More importantly, the new version introduces offline voice recognition capability—users can still use voice input smoothly even in weak network signal or no network environment, significantly improving the reliability of usage scenarios.

In terms of privacy, WeChat Input Method emphasizes strictly following the "minimum and necessary" principle, and all voice data processing is carried out in accordance with the "WeChat Input Method Privacy Policy." The official clearly stated that only the information necessary for the function to run is obtained, and no additional voice content is collected or uploaded. Especially in offline mode, the data is fully processed locally, without going online or being stored.

The release of this 3.0 version is not just a technical iteration, but also a clear signal of WeChat Input Method's transformation from a "typing tool" to an "intelligent voice entry point." In the current era where AI assistants and voice interaction are becoming core entry points of the mobile ecosystem, WeChat is trying to leverage its vast user base and deep understanding of the Chinese context to create a more natural, private, and inclusive next-generation input experience—ensuring that everyone, regardless of what they speak, can be accurately "heard."