https://github.com/brianprost/aws-transcribe-audio-processing-tool

Last synced: 8 months ago
JSON representation

README

# AWS Transcribe Audio Processing Tool

A Deno-based tool for transcribing audio files using AWS Transcribe and formatting the output into readable conversation format.

## Prerequisites

- Deno installed (version 2.0 or later)
- AWS credentials configured
- An S3 bucket with your audio files

## Setup

1. Clone the repository
2. Create a `.env` file with your AWS configuration:

```shell
BUCKET_NAME=your-bucket-name
```

## Usage

### Transcribing Audio

```bash
deno run --allow-read --allow-write --allow-env --allow-net main.ts your-audio-file.mp3
```

This will:

1. Upload and process your audio file using AWS Transcribe
2. Save the raw transcript as `your-audio-file.mp3_transcript.json`

### Formatting Transcript

To convert the JSON transcript into a readable format:

```bash
deno run --allow-read --allow-write parse.ts your-audio-file.mp3_transcript.json
```

This will generate a formatted text file with timestamps and speaker labels.

## Output Format

The formatted transcript will look like:

```plaintext
[0:00] Speaker 1:
Hello, welcome to the meeting.

[0:03] Speaker 2:
Thank you for having me.
```

## License

MIT