Timekettle W4 Plus AI Interpreting Earbuds Released: Bone Voiceprint Pickup Supports 54 Languages
Timekettle showcased the W4 Plus AI interpreting earbuds at the APEC Technology Exhibition in Shenzhen. The earbuds use a bone voiceprint sensor and a dual-microphone hybrid pickup solution to reduce ambient noise interference at the pickup level and improve translation accuracy. The product supports 54 languages and 108 accents, with 13 built-in offline language pairs, covering four major scenarios: two-person conversation, live listening, phone call translation, and audio/video translation. A single translation session lasts over 3 hours, and the device offers 12 hours of total battery life. Standard edition features are available permanently, while the Lite edition offers an unlimited-duration subscription at 98 yuan per month.
Body · AI translation
At the Shenzhen APEC tech exhibition, Timekettle showcased its W4 Plus AI interpreting earbuds, offering a noteworthy solution sample for the increasingly competitive translation earbud track. As nearly every audio brand piles AI translation features onto earbuds, consumers are gradually realizing a fact: there are many earbuds that can translate, but very few that are truly usable in noisy exhibition halls, airports, and at negotiation tables. The difference in translation capability often lies not in the large-model text translation stage, but in whether the earbuds can first cleanly pick up the speaker's voice. This is also why the W4 Plus chose to work on the underlying pickup layer.
The product adopts a bone-voiceprint sensor plus dual-microphone hybrid pickup solution: the bone-voiceprint sensor captures the signals generated by skull vibration when a person speaks, naturally isolated from airborne ambient noise, avoiding noise interference at the source; the dual microphones supplement the pickup of voice details such as breath and intonation, with the two sets of units each performing their own role to form a hybrid pickup system. Compared with noise-reduction solutions that simply stack microphones, this architecture can more stably extract effective human voice in complex sound fields where background noise and human voice overlap, reducing recognition misjudgments and thereby improving translation accuracy.
Pickup is the first step, and scenario coverage is another watershed. The W4 Plus supports 54 languages and 108 accents, with 13 built-in offline language pairs, and can still complete basic conversational translation in disconnected or weak-network environments, without binding the experience to the cloud. Centered on the real workflows of business users, the product connects four major scenarios: two-person conversation, live listening, phone-call translation, and audio-video translation: during face-to-face negotiations it provides two-way real-time interpreting; on transnational overseas calls, the other end of the call can complete two-way translation without any device; and online meetings can also be seamlessly connected. With features such as AI real-time summarization, domain-specific professional translation, and conversation memory, it can conveniently complete meeting records and terminology handling beyond translation, fitting the actual usage habits of foreign trade and cross-border business.
Lightweight design is another dimension that has been taken seriously. The charging case weighs 116g and can fit into a suit pocket; a single charge supports over 3 hours of translation and the whole device offers 12 hours of translation battery life, truly bringing "portable simultaneous interpreting" to a level that can be carried daily, rather than a device that needs to be specially packed for going out. On the underlying system side, the Babel OS 3.0 device-cloud integrated architecture allows local inference and cloud large models to work together, balancing response speed and translation quality.
The product also offers flexible choices for different users: the standard version's translation, AI, and offline features are all permanently available, suitable for high-frequency business users seeking a one-time purchase; the Lite version covers users with lighter, scenario-based usage through a free plan and a 98 yuan/month unlimited-duration subscription. This dual-plan design also reflects Timekettle's thinking on the tiered needs of translation earbud users—not all users need to pay for the full set of features.
Returning to the industry perspective, competition in the translation earbud track is entering deep waters. As large-model APIs become increasingly accessible, the threshold of "can it translate" has been leveled, and what truly creates a gap is pickup reliability, scenario completeness, and system refinement. As the pioneer of the translation earbud category, Timekettle has gone from creating two-way simultaneous interpreting in 2021 to today's W4 Plus iterating with a combination of bone-voiceprint pickup and a device-cloud system, following a route of continuous deep cultivation of a vertical category. Whether the W4 Plus can further consolidate its market position amid fierce competition is worth continued observation.