AutoShow vs ChapterGen

AutoShow is a free, MIT-licensed full-stack application for AI processing of audio and video. It accepts direct uploads and can extract audio from YouTube videos and streaming platforms. Transcription options include Groq Whisper, DeepInfra, YouTube Captions, and HappyScribe, with automatic speaker diarization. OpenAI GPT, Claude, and Google Gemini can produce episode descriptions, summaries, and chapters. The project also documents text-to-speech with OpenAI or ElevenLabs voices, generated cover images, and background music through ElevenLabs. It is available as a web interface, API, or self-hosted deployment, with Docker support and localhost:4321 as the documented local address. API endpoints cover health checks, jobs, models, processing, uploads, and media processing. Production documentation describes source-URL SSRF validation, USD cost estimates before jobs, structured JSON logs, and optional HTTP log forwarding. Google Drive OAuth setup remains manual.

ChapterGen generates YouTube chapters from uploaded video or audio files. It transcribes the audio, detects topic shifts, identifies sections, and creates chapter titles. Because it works from uploads, creators can prepare chapters for unpublished, private, or unlisted videos without providing a public YouTube URL. Audio-only material such as podcasts, interviews, and voice recordings is supported. Users can edit titles and timestamps before copying the output, which starts at 0:00 and is formatted for a YouTube video description. The FAQ lists MP4, MOV, AVI, WebM, MKV, MP3, WAV, M4A, AAC, OGG, and FLAC, with a maximum file size of 500 MB. The site's Whisper-based AI auto-detects language and supports 90+ languages. The homepage says uploads are deleted automatically after processing. The Free plan costs nothing and allows 2 videos of up to 20 minutes each, with no signup. Paid one-time plans are $5, paid once for 5 videos and $12, paid once for 15 videos. ChapterGen is a web tool.