thewh1teagle/vibe
Vibe
Vibe is a cross-platform transcription tool that converts spoken audio into text by running Whisper neural models directly on your device, with no cloud dependency. It can transcribe audio from files, microphones, system output, and network streams, and supports both batch processing of multiple files and real-time captioning from continuous input.
Beyond basic transcription, Vibe identifies and labels different speakers through speaker diarization, and offers a choice of Command-Line Interface or HTTP API for automated and remote workflows. It also includes plugins to export transcripts to common subtitle and document formats, and can summarize or translate transcripts using local or cloud AI models.
The tool combines local AI inference with flexible audio capture and output, making it suitable for a wide range of offline speech-to-text tasks. Documentation and installation instructions are available from the project repository.
Features
- Speech to Text Transcription - Transcribes spoken audio from files, microphones, and streams into text with precise timestamps.
- On-Device Transcriptions - Transcribes audio by running local neural models directly on-device without cloud dependencies.
- Audio Transcription - Converts spoken audio from files, microphones, or system output into text with accurate timestamps and speaker labels.
- CLI Transcription Tools - Transcribes audio files by running a command-line tool or making HTTP requests for automation.
- Real-Time Transcription - Produces live captions with low-latency updates from continuous audio input.
- Local Model Inference Servers - Runs speech-to-text entirely on-device by loading and executing neural network models without any cloud dependency.
- Real-Time Audio Transcribers - Transcribes audio from local files, system output, microphone, and online streams with live preview and caption generation.
- Transcription with Speaker Labels - Separates and labels speakers by voice characteristics to generate multi-speaker transcripts with precise timestamps.
- Whisper-Based Engines - Runs local Whisper models for offline speech-to-text with live captions and batch processing.
- Command-Line HTTP Clients - Offers both a CLI tool and an HTTP server for scriptable and remote transcription workflows.
- CLI and HTTP Server Interfaces - Provides both a command-line tool and an HTTP server offering the same transcription functionality for automation and integration.
- Live Captioning Integrations - Produces live captions with low-latency updates from continuous microphone or stream audio input.
- Batch Transcription - Processes multiple audio files sequentially or concurrently using a queued job manager for efficient batch transcription.
- HTTP API Servers - Offers both a command-line tool and an HTTP API for scriptable and remote transcription workflows.
- Transcription Exporters - Exports transcripts to common subtitle and document formats for use in other applications.
- Speaker Diarization - Separates and labels speakers by analyzing vocal characteristics and timing patterns across the audio.
- Batch Processing Pipelines - Orchestrates sequential or concurrent transcription of multiple audio files with queued job management.
- Audio Streaming Pipelines - Captures continuous audio from microphone or network streams and processes it in real-time for live caption output.
- Audio Source Captures - Ingests audio from files, microphones, system output, and network streams using platform-specific audio APIs.
- Audio Generation and Processing - All-in-one solution for efficient audio and video transcription.
- Media & Communication - Audio and video transcription tool.
- General Utilities - Tool for transcribing audio and video.
