US-based Deepgram Partners with Fortanix and NVIDIA to Shift On-Premise Voice AI to Confidential Computing Deployments
Recently, US-based real-time voice AI infrastructure company Deepgram partnered with data security f...
Microsoft's MAI-Transcribe-1.5 Integrates with Foundry, 43-Language Transcription Model Completes Voice AI Workflow
On June 2, Microsoft unveiled new members of the MAI model family during Build 2026, including MAI-T...
China's Soul App Open-Sources SoulX-Transcriber, a Multi-Person Conversation Transcription Model Integrating Speaker, Timestamp, and Text Recognition
On June 3rd, the Soul App AI team (Soul AI Lab), in collaboration with the ASLP@NPU team from Northw...
China's Tencent Cloud Partners with US-Based Soniox to Promote Multilingual Voice AI
Tencent Cloud has entered into a strategic partnership with Soniox, a voice AI company headquartered...
NTT Japan Launches Multimodal Explainable AI Reasoning Framework, Visual Language Models Enter Calibration Phase for Trustworthy Output
NTT Japan recently announced the establishment of the "Rationale-Enhanced Decoding" multimodal expla...
China's Tencent Cloud Partners with US-based Soniox to Integrate Real-Time Speech Transcription into Global Communication Infrastructure
China's Tencent Cloud recently entered into a strategic partnership with Soniox, a San Francisco-bas...
China's MiniMax Launches M3 Model, Pushing AI Competition Toward Long-Task Agents with Million-Token Context
On June 1, Chinese AI company MiniMax launched its next-generation general-purpose model, MiniMax M3...
China's DeepSeek-V4-Pro API Adjusted to 1/4 of Original Pricing, Long-Term Low-Price Strategy Drives Down Large Model Inference Costs
On May 22, DeepSeek's official pricing page showed that the API price for the DeepSeek-V4-Pro model ...
Germany's DFKI and RPTU Research Team Launch Privacy Guardrail 0.2.0, Safeguarding Sensitive Data in AI Chats with Local Anonymization
The German Artificial Intelligence Research Center (DFKI) has moved the protection of sensitive data...
Anthropic Welcomes OpenAI Founding Member Andrej Karpathy, Returning to the Forefront of Large Model R&D
On May 19 local time, OpenAI founding member Andrej Karpathy updated his personal status, announcing...
xAI Launches Skills Feature Across All Platforms in the U.S., Grok Gains Permanent Cross-Conversation Memory, Evolving Towards an Automated Workspace
Elon Musk's AI company xAI announced on May 18 local time the simultaneous launch of the "Skills" fe...
US-Based Coupa Acquires UK's Rossum to Strengthen AI Document Processing, T-LLM Replaces Traditional OCR to Cover End-to-End Spend Management
Autonomous spend management platform Coupa officially announced at its Inspire 2026 conference in La...
Tencent Yuanbao Upgrades Again, Adds Smart Summary and Analysis of WeChat Chat Records
Tencent's AI assistant Yuanbao announced a feature upgrade on May 13, officially supporting smart su...
OpenAI Launches GPT-Realtime Series of Three Audio Models, Integrating GPT-5-Level Reasoning into Voice Interaction for the First Time
OpenAI has officially launched the GPT-Realtime series, comprising three real-time audio models name...
Twilio Unveils Four New Conversational Layer Capabilities at SIGNAL Conference in San Francisco, Building Persistent Memory and Cross-Channel Orchestration for Customer Interactions
On May 6, 2026, U.S. cloud communications platform Twilio officially launched four new platform capa...