Audio transcription with Telegram and Groq Whisper
$20/month : Unlimited workflows
2500 executions/month
THE #1 IN WEB SCRAPING
Scrape any website without limits
HOSTINGER 🎉 Early Black Friday Deal
DISCOUNT 20% Try free
DISCOUNT 20%
Self-hosted n8n
Unlimited workflows - from $4.99/mo
#1 hub for scraping, AI & automation
6000+ actors - $5 credits/mo
Short description
Transcribe Telegram voice/audio messages to text using Groq’s OpenAI-compatible Whisper endpoint. Replies are delivered either as a Telegram message or as a downloadable .txt file, plug-and-play for n8n with minimal setup.
Who’s it for / Uses
- Educators, podcasters, interviewers, and support teams who need quick voice → text conversions.
- Automating meeting notes, voice feedback, voicemail transcription, or speech logging.
- Useful when you want transcripts pushed to chat or saved as files for archiving.
How it works (overview)
- Telegram Trigger — workflow starts on incoming message.
- Switch (Audio/Voice) — detects
voiceoraudio. If neither, replies “Wrong file type” and stops. - Telegram Download — downloads the audio using the
file_id, outputs file path/binary. - Set Node (credentials + options) — stores
Groq_APIandTelegram_access_token(required) andtranscript_output_format(messageorfile). - HTTP Request → Groq (Whisper) — uploads audio (multipart/form-data) to Groq’s transcription endpoint and receives text.
- Reply Switch — routes to either:
- Message branch: send transcribed text as a Telegram message.
- File branch: convert transcript to
.txtand send as a document.
Requirements
- n8n instance (cloud or self-hosted) with internet access
- Telegram bot token (create via BotFather)
- Groq API key (create at https://console.groq.com/keys)
- Basic n8n nodes: Telegram Trigger, Switch, Telegram Download, Set, HTTP Request, Convert to File, Telegram Send Message/Document
Important setup & notes
- Mandatory: Add
Groq_APIandTelegram_access_tokenin the Set node (or use n8n Credentials). The workflow will fail without them. - Do not hardcode keys in HTTP node fields that will be exported/shared. Use Set fields or n8n Credentials.
- Include sticky notes explaining each node (yellow note with full description recommended). Sticky notes should show setup steps and required fields.
- Before publishing: remove personal IDs and secrets, test with sample voice messages, and verify Groq response schema to map the transcript field correctly.
Security & best practices
- Use n8n Credentials or environment variables in production.
- Rotate API keys if they become exposed.
- Keep the Set node private when sharing templates; instruct users to replace keys with their own.