Search AI Products and News

NVIDIA's Canary-Qwen-2.5B sets a 5.63% WER record on Hugging Face OpenASR. This CC-BY licensed model combines FastConformer encoder with Qwen3-1.7B LLM decoder for efficient speech-to-text and NLP. Supports multi-GPU deployment for cloud/edge applications.....

Xiaomi Cars launches 'Quick Find' feature for YU7 models, using AI-powered image recognition to help locate parked cars via photos and voice queries.....

IndexTTS2 is an upcoming film-grade TTS model with 3 breakthroughs: 1) Local deployment & open weights; 2) Zero-shot voice cloning; 3) Pioneering emotion cloning. Supports precise duration control for dubbing (Chinese/English).....

According to Science and Technology Daily, the world's first generative artificial intelligence radiology system integrated into clinical processes developed by Northwestern University School of Medicine is revolutionizing medical imaging diagnostics. This system can identify life-threatening conditions within milliseconds, providing an innovative solution to the global shortage of radiologists. The AI system has been fully deployed in 12 hospitals affiliated with Northwestern University. In five months of actual application in 2024, the system successfully analyzed nearly 24,000 radiology reports, increasing report generation efficiency by an average of 15.5%, boosting work efficiency for some doctors.

Recently, Hume AI officially released its third-generation voice interaction model, EVI3. This new voice AI has drawn extensive attention in the industry due to its outstanding emotion understanding capabilities and personalized interaction experience. EVI3 can accurately identify emotions in users' voices and generate specific styles of sound and personality based on user preferences, marking a significant breakthrough in the field of emotional interaction and natural communication. Below is the latest information and in-depth analysis about EVI3 from AIbase. Experience it at: https://demo.

Recently, Qwen released CoGenAV, innovating speech recognition technology with the concept of audio-visual synchronization. It effectively addresses the challenge of noise interference in speech recognition. Traditional speech recognition performs poorly in noisy environments, while CoGenAV takes a different approach by learning the temporal alignment relationships among audio-visual-text, building a more robust and generalizable speech representation framework, systematically improving tasks such as speech recognition (VSR/AVSR), speech reconstruction (AVSS/AVSE), and audio-visual synchronization (A').

Nari Labs, a two-person startup, has released Dia, a 1.6 billion parameter text-to-speech (TTS) model designed to generate natural conversations directly from text prompts. Co-founder Toby Kim claims Dia surpasses proprietary offerings from competitors like ElevenLabs, as well as Google's NotebookLM AI podcast generation capabilities, and potentially even OpenAI's recently released gpt-4o-mini.

Traini, an AI-powered pet language translation app, has quickly gained popularity in the English mobile market since its launch. Developed in collaboration with animal behaviorists, Traini utilizes photo, video, and sound analysis to bridge the communication gap between humans and dogs, offering pet owners a revolutionary way to interact with their beloved companions.

From April 1st to 18th, approximately 100,000 ninth-grade students in Wuhan participated in their physical education entrance examination. This year's exam introduced a new AI intelligent judging system, marking a significant shift in physical examination assessment. Traditionally, physical education exams relied on manual timing and scoring. This year's exam utilized intelligent devices and cloud management to ensure fairness and efficiency. The exam incorporated several AI technologies, including facial recognition, electronic distance measurement, and wireless sensing, covering 14 items across three categories.

On March 13th, Sesame unveiled its latest speech synthesis model, CSM, attracting significant industry attention. According to the official introduction, CSM adopts an end-to-end Transformer-based multimodal learning architecture. It understands contextual information to generate natural and emotionally rich speech with stunningly realistic sound. The model supports real-time speech generation, processing both text and audio inputs. Users can also control features such as tone, intonation, rhythm, and emotion by adjusting parameters, showcasing high flexibility. CSM is considered a breakthrough in AI speech technology.

Baidu AI has announced the open-sourcing of PP-TableMagic, a new generation table recognition solution, marking a significant breakthrough in structured information extraction from tables. PP-TableMagic aims to address the limitations of traditional table recognition technologies in complex scenarios. Through an innovative multi-model network architecture, it achieves high-precision end-to-end table recognition and supports highly customizable model fine-tuning for all scenarios.

On March 11th, the Tongyi Lab team announced the open-sourcing of the R1-Omni model, marking a significant advancement in multimodal model development. This model integrates reinforcement learning with verifiable reward (RLVR) methods, focusing on improving reasoning capabilities and generalization performance in multimodal emotion recognition tasks. R1-Omni's training is divided into two stages. In the cold-start phase, the team fine-tuned the model using a combined dataset containing 580 video clips sourced from Explainable Multimodal Emotion...

The AI revolution in image recognition is rapidly evolving, moving beyond simple cat-dog classification. Current trends focus on much finer distinctions, such as identifying the year and model of a specific car or even subtle differences like the thickness of a bird's eyebrow. However, while neural networks excel at classification, explaining their reasoning ('Why did you classify this as X?') often proves challenging. Traditional Class Activation Maps (CAM)