Whishper

Overview

Whishper is an open-source, 100% local audio transcription and subtitling suite with a full-featured web UI. It offers various features such as transcribing any media to text, downloading transcriptions in multiple formats, translating transcriptions to different languages, and a powerful subtitle editor. Whishper is fast and can run on both CPU and GPU. It also has future plans to include features like local folder as media input, full-text search, user authentication, audio recording, and non-NVIDIA GPU support. The project is structured into three main pieces: Transcription-API, Whishper-Backend, and Whishper-Frontend.

Features

• Transcribe any media to text: Supports transcribing audio, video, and more.
• Download transcriptions in many formats: Offers TXT, JSON, VTT, SRT formats, and allows copying raw text to clipboard.
• Translate transcriptions: Can translate transcriptions to any language supported by Libretranslate.
• Powerful subtitle editor: Includes features like transcription highlighting, CPS warnings, segment splitting, segment insertion, and subtitle language selection.
• 100% Local: Transcription, translation, and editing happen locally on the user's machine.
• Fast: Uses FasterWhisper as the Whisper backend, resulting in faster transcription times, especially on CPUs.
• Quick and easy setup: Provides a quick start script and simplified setup steps.
• GPU support: Utilizes NVIDIA GPUs for even faster transcription times.
• CPU support: Can run on CPUs for users without GPUs.