whisper-srt 0.3.0

CLI tool to transcribe MP3 audio to SRT subtitles using OpenAI Whisper API

whisper-srt

CLI tool to transcribe MP3 audio files to SRT subtitle format using OpenAI Whisper API.

English | 日本語

Features

• Simple CLI for MP3 to SRT conversion
• High-accuracy transcription via OpenAI Whisper API (whisper-1)
• Built-in AI/tech terminology for better recognition
• Custom vocabulary support

Quick Start

No installation required - just run with uv:

# Set your OpenAI API key
export OPENAI_API_KEY="your-api-key"

# Transcribe (creates audio.srt)
uvx whisper-srt audio.mp3

Get your API key from OpenAI Platform.

Installation (Optional)

If you prefer a permanent install:

uv tool install whisper-srt
whisper-srt audio.mp3

Usage

whisper-srt input.mp3                    # Basic usage (default: English)
whisper-srt input.mp3 -o output.srt      # Specify output
whisper-srt input.mp3 --language ja      # Specify language (Japanese)
whisper-srt --help                       # See all options

Supported Languages

Default language is English (en).

Common language codes (ISO-639-1):

Code	Language	Code	Language
en	English	ja	Japanese
zh	Chinese	ko	Korean
es	Spanish	fr	French
de	German	pt	Portuguese

See OpenAI Whisper documentation for all supported languages.

Custom Vocabulary

Create ~/.config/whisper-srt/vocabulary.txt with domain-specific terms:

YouTube
Podcast
MyCompanyName

Pricing

$0.006/minute (~$0.36/hour)

Development

git clone https://github.com/tomada1114/whisper-srt.git
cd whisper-srt
uv sync --dev
make ci   # lint + type-check + test

License

MIT