agentsclimarketplace

Faster whisper srt

Skill allenphant/faster-whisper-SRT-converter/.agent/skills/faster-whisper-srt

Local audio/video to SRT subtitle converter powered by faster-whisper. Supports batch processing, 12 Whisper model sizes, real-time progress bar, and auto audio extraction from video. No API key needed.

Install
npx -y skills add allenphant/faster-whisper-SRT-converter --skill faster-whisper-srt

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

This skill should be used when the user asks to "convert audio to srt", "generate subtitles from audio/video", "create srt from mp3/wav/m4a/flac/mp4", "transcribe audio to subtitles", "把音訊轉成字幕", "產生 SRT 字幕檔", or needs to generate SRT subtitle files from audio or video files using the faster-whisper engine with model selection and progress display.

SKILL.md

2.7 KB, as published. Nobody here has run it

Faster Whisper SRT Converter

Purpose

Convert audio and video files to SRT subtitle files using the faster-whisper speech recognition engine. Supports multiple Whisper model sizes, real-time progress display, and automatic audio extraction from video files.

When to Use

  • User wants to convert audio files (MP3, WAV, M4A, FLAC, OGG, AAC, WMA) to SRT subtitles
  • User wants to extract subtitles from video files (MP4, MKV, AVI, MOV, WEBM, FLV)
  • User mentions "transcribe", "subtitles", "SRT", "字幕", "逐字稿"

Prerequisites

  • Python 3.10+
  • faster-whisper and tqdm packages (pip install -r requirements.txt)
  • FFmpeg (only required for video file input)

Usage

Step 1: Run the Script

Execute the conversion script with the target audio or video file:

python faster_whisper_srt.py <input_file> [--model MODEL] [--max-chars MAX_CHARS]

Step 2: Choose a Model (Optional)

Available models, from fastest to most accurate:

ModelSizeSpeedAccuracy
tiny~75 MBFastestLow
base~145 MBFastFair
small~490 MBMediumGood
medium~1.5 GBSlowHigh (default)
large-v3~3.1 GBSlowestHighest
large-v3-turbo~1.6 GBMediumHigh

Step 3: Check Output

The output SRT file will be saved in the same directory as the input file, named: <original_name>_<model>.srt

Examples

# Basic usage (default: medium model, 40 chars per line)
python faster_whisper_srt.py interview.mp3

# Use a fast model for testing
python faster_whisper_srt.py interview.mp3 --model tiny

# Use the best model for final output
python faster_whisper_srt.py interview.mp3 --model large-v3-turbo

# Shorter subtitle lines
python faster_whisper_srt.py interview.mp3 --max-chars 25

# Video file input (requires FFmpeg)
python faster_whisper_srt.py presentation.mp4 --model medium

Notes

  • First-time use of a model will trigger an automatic download. Subsequent runs use the cached model.
  • The script defaults to Chinese (zh) language detection. Modify the language parameter in the script for other languages.
  • Video processing requires FFmpeg to be installed and available in PATH.
  • Audio-only files (MP3, WAV, etc.) do NOT require FFmpeg.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.