AI-Video-Transcriber

Why we included this project

Most video transcription tools either force you to upload files or only handle one platform. This one accepts a URL from YouTube, TikTok, Bilibili, Apple Podcasts, SoundCloud, or 30+ other sources, and also lets you drag in local audio, video, or even plain text files. When a platform has native subtitles, it grabs those instantly and skips audio processing entirely, falling back to Faster-Whisper speech recognition only when no caption track exists. After transcription, the pipeline fixes typos, completes truncated sentences, and can produce summaries in a language of your choice, auto-translating when the source and target differ. You can point it at any OpenAI-compatible API endpoint from the UI, so it works with a local model or a hosted provider. That makes it a practical tool for anyone with a backlog of recordings, lectures, or meeting notes who wants clean, structured text instead of more hours of playback.