Apple announced that Apple Intelligence will officially launch in China this December with iOS 27.3, featuring core tools like writing assistance, photo cleanup, and real-time translation. Notably, the translation capability extends to AirPods (Pro 2 and later), enabling face-to-face cross-language communication for wearers. The feature requires specific hardware, including iPhone 15 Pro or newer models, and strict language settings.
Speech-to-Text
语音转文字
Technology that converts spoken language into written text
Curated Intelligence
Bee, the personal AI wearable maker acquired by Amazon, has launched two new devices: the wrist/pin-on Frontier ($74.99) and the phone-attached Eclipse ($99.99). Both feature dual microphones with noise filtering, long battery life, and come with redesigned companion apps. The pricing undercuts competitors like Plaud Note, targeting the mass market for personal voice recording.
XGIMI and DingTalk have launched the Q1, a bendable AI note-taking glasses priced from 1,299 RMB. The device features native integration with the DingTalk ecosystem to automate workflows from meeting recording and transcription to task creation, while utilizing lightweight titanium frames and head-gesture interactions to enhance wearability.
Developer Simon Willison demonstrated using the voice conversation mode in the ChatGPT desktop app's Codex tab to build a blog newsletter index page entirely through speech while cooking. This case validates the feasibility of a 'voice input + live preview' workflow for non-keyboard scenarios, highlighting the evolution of AI coding assistants from text commands to multimodal natural interaction.
Google has released an experimental note-taking app, AI Edge Foresight, which uses the on-device EmbeddingGemma 2 model to transcribe meetings and generate summaries entirely offline on macOS. This local-first approach addresses privacy concerns by eliminating cloud dependencies, though it currently supports only Apple Silicon devices.
Descript redesigned its translation pipeline to resolve the conflict between semantic fidelity and duration adherence in video dubbing using LLM reasoning. This approach automates traditionally expensive workflows, enabling high-quality, large-scale multilingual video localization.
Google announced upcoming releases of MedGemma 1.5 for medical image interpretation and MedASR for medical speech-to-text, but provided no technical specifications or performance data.