vibe
Transcribe on your own!
Vibe is an agent skill that enables AI assistants to transcribe audio and video files directly on the user's device using Whisper, Nemotron, and Parakeet models. The skill provides offline, privacy-focused transcription capabilities through an HTTP API with Swagger documentation, supporting multiple output formats including SRT, VTT, TXT, HTML, PDF, JSON, and DOCX. It integrates with AI platforms to offer batch processing, speaker diarization, translation, and transcript summarization using Claude API or local Ollama models. The skill is optimized for GPU acceleration across NVIDIA, AMD, and Intel hardware on Windows, macOS, and Linux.
Key Features
Use Cases
- 01AI agents automatically transcribing meeting recordings and generating summaries
- 02Content creation workflows where assistants extract and format subtitles from video files
- 03Multilingual document processing with translation to English and batch analysis
- 04Podcast and interview transcription with speaker identification for content indexing
- 05Accessibility automation generating captions for video content in multiple formats
- 06Research assistants processing audio datasets offline without privacy concerns
Related Skills
View moresuperpowers
An agentic skills framework & software development methodology that works.
skills
Skills for Real Engineers. Straight from my .agents directory.
skills
Public repository for Agent Skills
ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
vibe — FAQ
What is Vibe agent skill?+
Vibe is an agent skill that provides audio and video transcription capabilities through an HTTP API. It enables AI assistants to transcribe files locally using Whisper and other models, with support for batch processing, speaker diarization, and summarization.
How do I install Vibe for agent integration?+
Download and install the Vibe application from the official website, then enable the HTTP API server in settings. AI agents can connect to the local API endpoint to access transcription functionality via the Swagger-documented interface.
Which AI clients work with Vibe?+
Vibe provides an HTTP API that can be integrated with any AI agent or client capable of making HTTP requests. The skill works with Claude (via API integration), Ollama for local AI analysis, and custom agent frameworks through its REST interface.
Do I need API keys to use Vibe?+
Vibe runs fully offline without requiring API keys for basic transcription. Optional Claude API keys are needed only if you want to use the transcript summarization feature with Claude, while Ollama integration requires a local Ollama installation.
Is Vibe free to use?+
Yes, Vibe is open-source software available for free on GitHub. The core transcription functionality works completely offline at no cost, though optional Claude API summarization incurs standard Claude API fees.
What hardware does Vibe require?+
Vibe runs on Windows, macOS, and Linux with GPU acceleration support for NVIDIA, AMD, and Intel hardware via Vulkan and CoreML. While GPU acceleration improves performance, it can also run on CPU-only systems.
How do I install vibe?+
Open the source repository on GitHub and follow its README. vibe is a skill — MCP Agents Market links you directly to the official repo.
Is vibe free?+
vibe is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.