WhatsApp Video to Voice & Transcripts with Gemini | n8n

Turn WhatsApp Videos into Text—or Reply with AI Voice Notes—Automatically (n8n)

This n8n workflow listens for incoming WhatsApp messages, converts video audio to MP3, transcribes it with Google Gemini, and sends the transcript back to the sender. It can also generate voice notes from a /tts command using Fish Audio.

What this workflow does

• Triggers on new WhatsApp messages and extracts the sender number, message type, media URL, and text body.
• If the message includes video media:
  • Downloads the video from WhatsApp.
  • Creates a FreeConvert job to convert the media to MP3.
  • Uploads the video to FreeConvert, waits, then polls job status until conversion completes.
  • Downloads the resulting MP3, sets the audio MIME type, and transcribes the audio using Google Gemini.
  • Sends the transcription text back to the original sender via WhatsApp.
• If the message text contains /tts:
  • Calls a sub-workflow that generates speech with Fish Audio.
  • Uploads the audio to WhatsApp and returns it to the sender as an audio message.

Use cases

• Transcribe WhatsApp voice-style videos sent by customers into searchable text for support teams.
• Create instant multilingual-style voice replies: send /tts + your text and receive an audio response.
• Operational notes from WhatsApp: convert media to transcription for CRM or internal review (output is sent back to the user).

Technical details

• WhatsApp Cloud API: WhatsApp trigger + message upload/send steps (requires phone number ID and credentials).
• FreeConvert: video-to-MP3 conversion (requires FreeConvert API key; template currently uses a hard-coded Authorization header).
• Google Gemini (Google PaLM): audio transcription.
• Fish Audio: text-to-speech for /tts via a sub-workflow.
• Workflow control uses nodes including if, set, code, wait, and supporting steps (e.g., sticky note).

Ready to deploy: add WhatsApp, FreeConvert, Google Gemini, and Fish Audio credentials, configure your voice ID for TTS, then enable the workflow.