Search Results for "signal processing"

WhisperJAV is an open-source speech transcription pipeline designed specifically for generating subtitles for Japanese adult video content. The project addresses challenges that standard speech recognition models face when transcribing this type of audio, which often includes low signal-to-noise ratios and large numbers of non-verbal vocalizations. Traditional automatic speech recognition systems can misinterpret these sounds as words, leading to inaccurate transcripts. WhisperJAV introduces...

AudioMuse-AI is an open-source system designed to automatically generate playlists and analyze music libraries using artificial intelligence and audio signal processing techniques. The platform runs locally in a Dockerized environment and performs detailed sonic analysis on audio files to understand characteristics such as tempo, mood, and acoustic similarity. By analyzing the underlying audio content rather than relying on external metadata services, the system can organize large personal music libraries and generate curated playlists for different moods or listening contexts. ...

...It has 284B total parameters with 13B activated and supports a 1M-token context window, making it suitable for long-document reasoning, complex coding, agentic workflows, and large-scale information processing. The model uses a hybrid attention architecture that combines Compressed Sparse Attention and Heavily Compressed Attention to improve long-context efficiency, while Manifold-Constrained Hyper-Connections strengthen signal stability across layers. It is trained on more than 32T tokens and refined through a post-training pipeline that includes supervised fine-tuning, reinforcement learning, domain-specific expert cultivation, and on-policy distillation. ...