🎙️ youtube-dubber

Dub any YouTube video into Hindi (and 19 other languages) with a casual, engaging neural voice — like a desi YouTuber, not a robotic narrator.

Give it a URL → it pulls the captions, translates them to natural casual speech, generates a neural voice, and saves synced dubbed audio clips + a manifest. Fully headless — use it in scripts, servers, or pipelines.

⚡ Quick start

pip install youtube-dubber
export GROQ_API_KEY=gsk_xxxxxxxx        # free key → https://console.groq.com/keys

from youtube_dubber import dub

manifest = dub(
    "https://www.youtube.com/watch?v=VIDEO_ID",
    lang="hindi",       # any of 20 languages
    gender="female",    # "male" or "female"
    out="./out",
)
print(len(manifest["segments"]), "segments dubbed")

Or from the command line:

youtube-dubber --url https://youtu.be/VIDEO_ID --lang hindi --gender female --out ./out

Output: out/audio/<videoId>_<lang>_<gender>/seg_NNNNN.mp3 clips + out/manifest.json.

⚠️ System Requirements

pip install brings the Python code, but you must also have these on your system (they are not pip-installable Python packages):

💡 Install yt-dlp alongside the package: pip install "youtube-dubber[ytdlp]" ffmpeg must be on your PATH — it is a system binary, not a pip package.

🌍 Supported languages (20)

Indian (casual Hinglish / code-switch style — tech terms stay English): Hindi · Tamil · Telugu · Bengali · Marathi · Gujarati · Kannada · Malayalam · Punjabi · Urdu

International (full translation): Spanish · French · German · Japanese · Chinese · Korean · Arabic · Portuguese · Russian · Italian

Each has a male & female neural voice. Source language is auto-detected — dub an Arabic, Chinese, or English video into Hindi, all the same way.

🧠 Live progress (callback API)

from youtube_dubber import Dubber

def on_event(ev):
    if ev["type"] == "progress":
        print(f"[{ev['step']}] {ev['pct']}%  {ev['msg']}")
    elif ev["type"] == "segment":
        print("dubbed:", ev["dubbed"])

Dubber(lang="hindi", gender="male", out_dir="./out", on_event=on_event).run(url)

Event types: stream_url · progress · segment · done · error

✨ Why it's different

• Casual, not robotic — prompt-tuned to sound like a real creator ("तो भाई, scene ये है…"), not a textbook
• Auto language detection — works on any source language out of the box
• Disk cache — re-running the same video is instant (no re-translation/TTS)
• Rate-limit resilient — automatic retries so segments never silently vanish
• Fit-to-window timing — clips are speed-matched (≤1.4×) so dubbing stays in sync without sounding sped-up

🖥️ Want the desktop app instead?

This package is the headless engine. There's also a full Electron desktop app (paste a URL, watch the dubbed video live with subtitles) and a real-time Live Dub mode — see the GitHub repo:

👉 github.com/Ashut90/youtube-dubber

📄 License

GPL-3.0 — © 2026 Ashutosh (@Ashut90). Forks must stay open source and credit the original. See LICENSE.

Built with Groq · edge-tts · yt-dlp · ffmpeg.