https://github.com/zdwolfe/transcription-tools

Docker video transcriber, wrapper around OpenAI
https://github.com/zdwolfe/transcription-tools

openai transcription whisper whisper-ai

Last synced: 6 days ago
JSON representation

Docker video transcriber, wrapper around OpenAI

• Host: GitHub
• URL: https://github.com/zdwolfe/transcription-tools
• Owner: zdwolfe
• Created: 2024-09-15T18:15:30.000Z (2 months ago)
• Default Branch: main
• Last Pushed: 2024-10-06T17:45:37.000Z (about 1 month ago)
• Last Synced: 2024-10-18T06:53:35.630Z (28 days ago)
• Topics: openai, transcription, whisper, whisper-ai
• Language: Python
• Homepage:
• Size: 8.79 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Transcription Tools

A Docker tool for transcribing video files with spoken words. The intended use of transcriptions is as input to other text analysis tools, such as LLM summarization or indexing.

This is mostly a Docker wrapper around [OpenAI's Whisper](https://openai.com/index/whisper/).

## Features

- Can handle multiple video files, concatenating videos in order

- Extracts audio from video files

- Transcribes audio using [OpenAI's Whisper](https://openai.com/index/whisper/), output is Whisper transcript format.

- Outputs transcription in JSON and plain text formats

- Generates a manifest file with metadata and checksums

- GPU acceleration

## Requirements

- Docker

## Building the Docker Image

To build the Docker image, run:

```bash

docker build -t transcription-tools-gpu ./transcription

```

## Usage

```bash

docker run --gpus all -ti -v "$PWD/transcription/data:/usr/data" transcription-tools-gpu --input /usr/data/video.mkv  --output-dir /usr/data --model large-v3

```

Will output:

1. ``data/video1.manifest.json``

2. ``data/video1.transcript.json``

3. ``data/video1.transcript.txt``