Creators, researchers, and translators often need the same handful of things from a video: the audio, a short clip, a still, and a usable transcript. Most tools send the file somewhere else to do that work. YT Video Downloader + Transcriber keeps it on your machine.

It is a straightforward Windows desktop app. Paste a YouTube video ID, or open a local media file, then check the jobs you want. The app runs them in order with yt-dlp, ffmpeg, and Faster-Whisper. Finished files land in your Downloads folder.

Why we built it

Cloud transcription is convenient until the file is sensitive, the connection is poor, or you simply do not want another account in the loop. We wanted a tool that is local by default, honest about what it can do, and boring in the best way: pick a task, wait for the green light, click through to the file.

1. YouTube mode — one ID, several jobs

You do not paste a full URL. You paste the video ID (the part after v=), for example mkLjMu8SKqk. A preview thumbnail appears so you can confirm you have the right clip.

Then you choose any combination of:

• Download Transcript — speech-to-text on your PC
• Download Audio (MP3) — best available audio, converted to 192 kbps MP3
• Download Video — prefers MP4 up to 1080p, with fallbacks to WEBM, MKV, or AVI
• Extract Clip — start time as MM:SS plus duration in seconds, output as H.264/AAC
• Download Thumbnail — the YouTube thumbnail, or a specific frame if you set a timecode

Tasks run one after another. Status lights turn green when a job succeeds. Click a light to open that file in Explorer.

2. Local file mode — the same toolkit without YouTube

Switch the source to a video or audio file already on disk. You can transcribe it, cut a clip, or grab a thumbnail frame. YouTube download options hide themselves in this mode, so the window stays focused on the file in front of you.

That is the difference between a YouTube utility and a small media workshop. Lecture recordings, interviews, and client files can go through the same clip-and-transcript path.

3. Transcription that never leaves the machine

YT Video Downloader + Transcriber uses Faster-Whisper, not a hosted API. The default model is base. The first run downloads the model and caches it. After that, audio is decoded on your computer.

If a working NVIDIA GPU and CUDA libraries are present, the app remembers CUDA and uses it. If they are not, it falls back to CPU and remembers that choice. You can still force a retry later. Larger models are more accurate and slower; on CPU, base or small is a practical starting point.

Nothing is uploaded. Temporary audio used for transcription is cleaned up when the job finishes.

4. Resilient downloads, not brittle format IDs

YouTube changes formats often. The app does not pin a single format code and hope. Video prefers an MP4 pair up to 1080p, then any video-plus-audio pair, then a single best file. Audio prefers the best source and converts to MP3. If only WebM is available, you get WebM instead of a failed job.

Keep yt-dlp.exe current. File → Settings lets you point at an existing executable or download the latest nightly build. That path is saved under %APPDATA%\YTranscriber. An old extractor is the usual reason a download that worked last month suddenly fails.

5. Built for Windows, and for the way people actually work

The current UI and settings paths are written for Windows. You need Python, ffmpeg on PATH, a standalone yt-dlp.exe, and a JavaScript runtime (Node.js is the usual choice) so yt-dlp can solve YouTube’s player challenges. The in-app log panel is there on purpose: when something fails, it usually says why.

Use it only for content you have the right to download or transcribe. Respect YouTube’s terms and copyright.

Who it is for

• Content creators who need an MP3, a 20-second clip, and a thumbnail from the same video
• Researchers and students who want a local transcript of a talk they are allowed to keep
• Translators and editors who start from speech rather than a published subtitle file
• Anyone who prefers a Downloads folder to another cloud inbox

Try it

YT Video Downloader + Transcriber is an AppHaven desktop tool. If you want the longer walkthrough — interface, troubleshooting, and legal notes — see our YT Video Downloader + Transcriber page. For a license, a build, or a custom variant, write to projects@apphaven.net.