⚠️ For educational purposes only (or for free music) Please don't use this for illegal stuff. It's against Deezer's terms of service.

Your native language is not in the langs.json file ? Just make a pull request or send me a message !

• Get an arl cookie on Deezer for DEEZER_TOKEN ( see this repo)
• Create a bot on Telegram and grab a token with Bot Father (TELEGRAM_TOKEN)
• Activate Inline Mode on BotFather for the bot you just created

Search for music in inline mode :

@xxxxxxx_bot (album|track) <search>

Or send a Deezer, YouTube or SoundCloud link directly.

• Docker (Linux installation and Linux post-installation steps)
• Docker compose (should be included in the Docker installation)

Create a token.env file with the following content, replacing the values with your own tokens :

DEEZER_TOKEN=abcdefghijklmnoxxxxxxxxxxxx
TELEGRAM_TOKEN=123456789:ABCDEFGHIJKLMNOPQRSTUVWXYZ
BOT_LANG=fr

./update.sh

Or manually : git pull && docker-compose up -d --build

• Add DEEZER_TOKEN and TELEGRAM_TOKEN as variable environment
• python3.13 -m pip install -r requirements.txt
• python3.13 main.py

You should use a venv to avoid conflicts with your system python packages

The bot can bundle downloaded tracks into a zip archive with the cover art included.

In the token.env file, add the following line :

FORMAT=zip

If the zip file is too big (more than 50MB), the bot will split the zip file into multiple parts.

You can also make the bot move a file to a specific directory and send you the download link.

In the token.env file, add the following line :

FORMAT=zip
COPY_FILES_PATH=/path/to/your/directory
FILE_LINK_TEMPLATE=https://yourdomain.com/download/{0}

The {0} will be replaced by the file name.

I recommend using a reverse proxy to serve the files, like Nginx or Caddy.

If you use Docker, you can mount a volume to the container and set the COPY_FILES_PATH to /files, like this :

volumes:
  - /path/to/your/directory:/files

If you have Deezer premium, you can allow the bot to download FLAC files by adding the following line in the token.env file :

ENABLE_FLAC=1

By default, the bot sends the album cover along with links to the album and track. To disable this, set SEND_ALBUM_COVER to false :

SEND_ALBUM_COVER=false

YouTube has deployed multiple layers of protection that affect yt-dlp:

1. JS challenges (sig/nsig) — stream URLs are encrypted by obfuscated JS in the YouTube player, changing with each update.
2. poToken (BotGuard) — an attestation token proving the request comes from a real client (browser/mobile app).
3. Datacenter IP blocking — YouTube detects server IPs and requires login ("Sign in to confirm you’re not a bot").

By default, yt-dlp uses JSInterp (a Python-based JS interpreter) which is deprecated and no longer follows YouTube changes. It still works for some very popular videos, but fails with LOGIN_REQUIRED on many videos from datacenter IPs.

Using yt-dlp with EJS (a real JS runtime like Node/Deno/Bun) resolves the JS challenges but does not solve datacenter IP blocking. Without cookies or a residential proxy, many videos won’t work from a server.

SoundCloud restricts full track downloads to Go+ subscribers. Without authentication, yt-dlp only retrieves ~30-second previews.

Add a cookies.txt in ./local_resources. To generate this file, please see: https://github.com/yt-dlp/yt-dlp/wiki/FAQ#how-do-i-pass-cookies-to-yt-dlp.

You don’t need to restart the bot, it will automatically reload the cookies when downloading a YouTube video.

⚠️ Please note that your account may be banned if you use this feature.

# Global configuration
BOT_LANG=en

# Tokens
DEEZER_TOKEN=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
TELEGRAM_TOKEN=1234567890:ABCDEFghijklmnopqrstuvwxyz012345678

# Specific configuration
COPY_FILES_PATH=/files
FILE_LINK_TEMPLATE=https://example.com/dl/{0}
FORMAT=zip

docker logs telegramusic 2>/dev/null | grep "USER_DEBUG"

grep "USER_DEBUG"                     # All user activities
grep "USER_DEBUG.*user_id=123456789"  # Activities for specific user ID
grep "USER_DEBUG.*username=john"      # Activities for specific username
grep "USER_DEBUG.*Error"              # All user-related errors