Projects sharing features with BiliNote

VoiceInk is a system-wide speech-to-text dictation tool that converts spoken audio into text using local or cloud AI models. It functions as a local AI transcription engine and a context-aware voice assistant, allowing users to insert transcribed text directly into any active application on the operating system. The project distinguishes itself through the use of custom vocabulary management, which trains transcription engines to recognize industry-specific technical terms, professional terminology, and personal names. It further enhances output by using large language models to refine raw tr

Vosk is an offline speech-to-text engine and API that converts spoken audio into text locally on a device. It provides a cross-platform speech toolkit with language bindings for integrating voice recognition into server environments, Android, iOS, and Raspberry Pi. The project includes a speaker identification tool to distinguish between different voices and an acoustic model trainer for building custom neural network models. These training tools enable speech feature extraction and model accuracy evaluation to improve recognition for specialized domains. The system supports real-time audio

Mediago is a multi-platform video downloader and extraction utility that supports content retrieval from over a thousand different social and video sharing platforms. It functions as an AI-powered download automator that uses natural language commands to trigger and manage video download workflows. The project provides specialized capabilities for sniffing and capturing m3u8 and HLS video streams via a browser extension that detects resources on active web pages. It can be deployed as a containerized headless server, allowing users to manage remote download tasks across a network through a we

DeepSpeech is an open-source speech-to-text framework and machine learning engine designed to convert spoken audio into written text locally on a device. It provides on-device speech recognition that operates without requiring an internet connection to external servers. The system supports real-time speech transcription across a variety of hardware platforms, ranging from single-board computers and edge devices to GPU servers. This allows for audio analysis and processing directly on the local hardware.

Auto-subs is an AI transcription and automatic captioning tool that converts spoken audio from video files into synchronized subtitles. It functions as a subtitle generator and a transcription bridge, enabling the conversion of speech to text with automatic speaker identification and multi-language translation support. The software prioritizes data privacy by utilizing on-device AI inference to process audio and video files locally on the user's hardware. It distinguishes itself by offering deep integration with professional video editing workflows, allowing users to export timing and transcr

This project is a Telegram cloud mirror bot designed to download files from various internet sources and mirror them to Telegram or cloud storage. It functions as a specialized service for remote media downloading, torrent and Usenet leeching, and automated content mirroring. The bot distinguishes itself through deep integration with the Rclone ecosystem for managing, cloning, and migrating files across multiple cloud providers. It includes an RSS content automator to trigger downloads based on user-defined filters and utilizes service account rotation to bypass API quotas when managing cloud

Ecoute is a live transcription tool that provides real-time transcripts for both the user's microphone input (You) and the user's speakers output (Speaker) in a textbox.

Persepolis is a Python-based download manager and graphical user interface for the aria2 engine. It functions as a concurrent file downloader that splits files into multiple chunks to be retrieved simultaneously, maximizing available network bandwidth. The application integrates a media extraction engine to resolve direct URLs, enabling the retrieval of video and audio content from various online streaming platforms and hosting services. It provides tools for download queue management and automation, including the ability to schedule specific start times or define time windows for downloads

Streamlink is a command line video stream extractor that retrieves direct stream URLs from online services for use in external media players. It functions as a local media stream pipe, redirecting raw video data from web services into local files or players via standard input or HTTP. The project includes a headless browser stream scraper to intercept network requests and extract media data from script-heavy websites, alongside a dedicated processor for HLS and DASH segmented media streams. The tool utilizes a modular video plugin framework, allowing support for new streaming platforms to be

This project is a high-throughput transcription engine and PyTorch inference wrapper designed to convert spoken audio files into text using the OpenAI Whisper model. It functions as a hardware-accelerated speech-to-text transcriber that runs locally on a user's machine. The system focuses on AI model performance tuning to maximize hardware throughput. It utilizes GPU acceleration, half-precision floating point tensors, and Flash-Attention to reduce processing time and memory overhead during transcription. The implementation covers large-scale transcription workflows and local speech-to-text

Sherpa-ONNX is an ONNX-based speech processing toolkit that provides a local speech recognition engine, an on-device voice synthesis tool, and a speaker identification framework. It is designed as a cross-platform speech API that enables speech-to-text, text-to-speech, and speaker verification tasks to be executed locally on a device without requiring network access. The project is distinguished by its ability to perform zero-shot voice cloning and speaker diarization on-device. It supports a wide range of hardware accelerations, including GPU and various NPU architectures, and provides a Web

AI-Media2Doc is a self-hosted media transcription and content repurposing platform. It converts audio and video files into structured documents, searchable transcripts, and notes using a local speech-to-text pipeline and language model integration. The system enables the creation of illustrated articles by extracting video frames based on subtitle timestamps. Users can control the style and structure of generated text through custom behavioral prompts and integrate external language models via configurable API endpoints. The platform covers automated speech transcription, multimedia content

Livestreamer is a command-line stream extractor and modular streaming client designed to resolve hidden media source URLs from online services. It functions as a media pipeline utility that extracts direct media URLs and binary data for local saving, specialized processing, or routing into external media players. The project uses a plugin-based system with custom heuristic logic to extend service support and override default capabilities. It includes a network proxy router to bypass regional blocks and network restrictions through configurable HTTP and HTTPS proxies, and manages authenticatio

Tartube is a desktop video manager and web video downloader designed for capturing and storing video content from a wide variety of hosting sites for offline viewing. It serves as a graphical interface for the yt-dlp engine to download and organize videos from platforms such as YouTube and Twitch. The application enables web video archiving to prevent data loss from deletions or platform changes. It provides a dedicated interface for categorizing and managing large collections of downloaded web videos within an organized library. The system handles multi-platform video downloading and offlin

Cheetah is an LLM technical interview assistant composed of a native macOS application and a browser extension. It provides real-time coding and answering suggestions during technical interviews by combining live audio transcription with web-based context extraction. The system functions as a real-time interview coach that converts spoken questions into text using on-device speech-to-text processing. It uses a browser-integrated DOM scraper to extract live code and console logs, allowing the AI to analyze the current coding state and generate technical solutions based on the specific environm

Voicebox is a local speech processing system that provides text-to-speech generation, speech-to-text transcription, and voice cloning. It utilizes local machine learning inference and GPU acceleration to process audio and text data without relying on external API calls. The project features a voice cloning toolkit for creating synthetic profiles from audio samples and a timeline-based voice editor for composing multi-character conversations. It also includes an AI voice management API that allows external applications and AI agents to programmatically manage voice profiles and generate speech

node-ytdl-core is a JavaScript library for Node.js designed to extract metadata and stream video and audio content from YouTube. It serves as a media downloader and stream fetcher, allowing users to retrieve video details and media data from remote sources. The library provides specialized capabilities for video extraction, including the ability to parse media URLs for unique identifiers and analyze available formats. It allows for the selection and filtering of specific video and audio streams based on quality and resolution criteria. The project manages network traffic through rate limit a

RealtimeSTT is a local speech-to-text engine and real-time automatic speech recognition server. It utilizes transformer-based recognition and omnilingual pipelines to convert live audio streams into text, providing a WebSocket-based streaming API for raw PCM audio transmission. The project is distinguished by a dual-backend transcription pipeline that uses a lightweight engine for immediate partial suggestions and a heavier model for final high-accuracy results. It includes a wake word detection system to trigger recording and employs a shared-resource inference model to distribute heavy spee

This project is a command-line utility and Python-based scraper designed to download and archive educational media from Coursera for offline access. It functions as a tool for saving course syllabi, videos, subtitles, and notebooks to a local file system to ensure a secure backup of academic materials. The tool distinguishes itself by providing capabilities for course quiz extraction, retrieving questions and answers from both graded and ungraded quizzes. It supports the retrieval of learning materials from both legacy and on-demand platforms and implements interrupted download resumption to

This project is a knowledge base plugin and RAG context manager that uses a local vector database interface to enable semantic search and relationship mapping. It transforms text into numerical vectors to find semantically related notes and excerpts based on conceptual meaning rather than keyword matches. The system differentiates itself through a semantic graph visualizer that maps notes into clusters to reveal conceptual connections. It also features a context manager capable of bundling local notes and excerpts into reusable packs to provide grounded factual bases for large language model

DrQA is an open-domain question answering system that retrieves relevant documents from a large corpus and extracts specific answers to natural language questions. It is implemented as a neural network system that combines a document retrieval engine with a machine reading comprehension model. The system utilizes a two-stage pipeline architecture. A coarse-grained document retriever uses weighted word vectors to identify potential documents, while a fine-grained machine reading comprehension model identifies and extracts the exact text span containing the answer. The project also includes a

Kotaemon is an orchestration framework designed for building modular, agentic workflows that integrate document processing, retrieval-augmented generation, and multi-step reasoning. It provides a comprehensive platform for developing document-based question answering systems, allowing users to chain language models, prompt templates, and external tools into complex, automated pipelines. The system distinguishes itself through a highly modular architecture that emphasizes component-based composition and schema-driven data exchange. It supports autonomous agents capable of decomposing complex q

DeepPavlov is a conversational AI framework and deep learning NLP library designed for building end-to-end dialogue systems and chatbots. It functions as an NLP pipeline orchestrator that allows users to compose pre-trained models and text processing components into sequential data flows for complex linguistic tasks. The system is distinguished by its ability to act as a chatbot deployment server, exposing trained conversational models as web services via REST and Socket APIs. It utilizes JSON-based pipeline configurations and dynamic variable interpolation to decouple model logic from infras

Paper-qa is a retrieval augmented generation system designed for question answering and analysis of scientific literature and technical documents. It functions as an LLM-powered research assistant that extracts grounded answers and summaries with citations from a document library. The system utilizes an agentic RAG orchestrator to iteratively refine search queries and gather evidence through multi-step tool calling. It features a multimodal document parser that extracts text, tables, and images from PDFs, alongside a vector-based indexer that embeds and caches document libraries for efficient

fun-rec is a learning guide and framework for building personalized recommendation systems, covering everything from deep learning ranking to generative recommendation paradigms. It provides instructional content on constructing industrial-grade architectures that span offline data processing and real-time online serving. The project distinguishes itself by focusing on generative recommendation, treating the suggestion process as a sequence-to-sequence task using large language models and transformer models to generate item identifiers rather than traditional ranking lists. It also emphasizes