Introduction to Qianwen Input Method
Qianwen Input Method is launched by Alibaba's Tongyi Qianwen team as an AI input method, built on the Tongyi Qianwen large model and CosyVoice speech capabilities, supporting both voice and keyboard input modes. The product converts spoken language into neat written text in real time, automatically removing filler words, correcting slips of the tongue, and intelligently polishing the text, while also supporting 9 dialects including Cantonese and Sichuanese as well as mixed Chinese-English speech.
Main features of Qianwen Input Method
-
AI voice input, speak and it becomes text: Based on the Qianwen speech large model, it supports fast speech, soft speech, dialect accents, mixed Chinese-English speech, and weak network environments, automatically removing filler words, smoothing word order, and paragraphing, and can break content into lists, minutes, or notes.
-
Multiple ways to speak: Supports speaking by tapping the microphone or long-pressing the spacebar, swiping left to undo and swiping right to send directly, making operation flexible and handy.
-
9 dialects, understood as soon as you speak: Supports 9 dialects including Cantonese, Sichuanese, and Shanghainese, with no need to switch manually—whatever you say is what comes out.
-
Voice can also write long content: Supports recording for up to 3 minutes at a time, so long emails, weekly reports, and meeting points can be used directly once spoken.
-
Smooth keyboard, also great for typing: Supports switching between 26-key and 9-key layouts, with intelligent prediction, fuzzy pinyin, input error correction, and one-handed mode, so fast typing mistakes can still produce the right words.
-
Pure and ad-free, no login required: No ads, no pop-ups, no news feed, and all core features can be used without registration.
