Generate a music press release draft for a new single with OpenAI GPT-4o and Google Docs

Description

Description

This workflow automates the creation of press releases for music artists releasing a new single. Upload your MP3, fill in basic info, and receive a publication-ready press release saved as a Google Doc — powered by multiple OpenAI models working in sequence.

The system analyzes your audio, extracts or transcribes lyrics, researches your artist background, interprets the song's themes, and writes a polished press release in your chosen language.

Key Features:

Automatic lyrics handling**: Upload a .txt file or let Whisper transcribe from audio
AI audio analysis**: GPT-4o Audio Preview identifies genre, mood, BPM, key, and instrumentation
Smart artist research**: If no bio is provided, GPT-4o Search Preview researches the artist online
Deep song interpretation**: AI creates a creative brief analyzing themes, imagery, and press angles
Multi-language support**: Press release written in English, Czech, German, Spanish, French, or other
Google Docs output**: Final press release saved as an editable Google Doc
Email delivery**: Artist receives an email with a direct link to the document

Perfect For:

Independent musicians and bands releasing new music
Record labels and artist managers
PR agencies in the music industry
Music publicists and booking agents
Anyone who needs a quick, professional press release draft

Use Cases:

New single / EP / album announcement
Press kit preparation
Music blog and magazine outreach
Festival and playlist submission cover letters
Social media press content drafts

How It Works:

Step 1 — Form Submission
Artist fills out a web form with:
Audio file (.mp3, required)
Lyrics file (.txt, optional)
Artist name, single name, release date
Artist bio (optional)
Preferred language for the press release

Step 2 — Lyrics Extraction
If a lyrics .txt file is uploaded → text is extracted directly
If not → audio is transcribed using OpenAI Whisper API (model: whisper-1)

Step 3 — Audio Analysis
File size check (max 25 MB)
Audio converted to Base64
Sent to GPT-4o Audio Preview which returns structured JSON: genre, subgenre, mood, instrumentation, energy level, and overall description

Step 4 — Artist Context
If band bio was provided in the form → used directly
If not → GPT-4o Search Preview researches the artist online
If no info found → fallback text is used ("emerging artist")

Step 5 — Deep Song Interpretation
GPT-4o creates a creative brief combining lyrics, audio analysis, and artist context. Output includes: core theme, emotional arc, imagery, unique angles, and press coverage suggestions.

Step 6 — Press Release Writing
GPT-4o writes a publication-ready press release with: headline, dateline, 3 body paragraphs, artist quote placeholder, boilerplate, and contact info — all in the selected language.

Step 7 — Output & Delivery
A new Google Doc is created with the press release text

Required Credentials:

| Credential | Type | Used By |
|-----------|------|---------|
| OpenAI API Key | Header Auth (Authorization: Bearer sk-...) | Whisper, Audio Analysis, Research, Interpretation, Press Release |
| Google Docs | OAuth2 | Create document + Insert text |

Setup Instructions:

Import the workflow into your n8n instance
Create OpenAI credential: Go to Credentials → Add → Header Auth → Name: Authorization, Value: Bearer sk-your-key
Create Google Docs OAuth2 credential: Follow n8n's Google OAuth2 setup guide
Create Gmail OAuth2 credential: Same Google OAuth2 flow with Gmail scopes
Assign credentials to all relevant nodes (5× OpenAI, 2× Google Docs, 1× Gmail)
Add a Gmail node at the end of the workflow (after "Insert Press Release Text") if not already present
Activate the workflow
Share the form URL with your artists

Configuration Notes:

Language dropdown**: Modify the Form Trigger to add more language options if needed
Audio size limit**: Currently set to 25 MB (OpenAI Whisper limit). Adjustable in the "Check Audio Size" code node
Max tokens**: Interpretation and Press Release steps use 2048 tokens. Increase if you need longer outputs
Artist quote**: The press release includes a placeholder [ARTIST QUOTE: "..."] — the artist should replace this before distribution

Cost Breakdown:

API Costs (Per Press Release)

| Service | Model | Est. Cost |
|---------|-------|-----------|
| Whisper (if no lyrics file) | whisper-1 | ~$0.006/min of audio |
| Audio Analysis | gpt-4o-audio-preview | ~$0.02–0.05 |
| Artist Research (if no bio) | gpt-4o-search-preview | ~$0.01–0.03 |
| Song Interpretation | gpt-4o | ~$0.01–0.03 |
| Press Release Writing | gpt-4o | ~$0.01–0.03 |
| Google Docs | — | Free |

Estimated total per press release: ~$0.05–0.15

Contact:

📧 Email: [email protected]
💼 LinkedIn: https://www.linkedin.com/in/vaclavcikl/