Podcastfy Openclaw Skill

Podcastfy Skill for OpenClaw

> Transform content into engaging AI-generated podcasts using Podcastfy.

Overview

Podcastfy is an open-source Python package that transforms multi-modal content (text, images, websites, PDFs, YouTube videos) into engaging, multi-lingual audio conversations using GenAI.

It's like NotebookLM's podcast feature but open-source and programmatic.

Prerequisites

1. Install Podcastfy

pip install podcastfy

2. Install FFmpeg (required for audio processing)

# macOS
brew install ffmpeg
Ubuntu/Debian
sudo apt install ffmpeg

3. Configure API Keys

Create a .env file with your API keys:

# Required: At least one LLM + one TTS
GEMINI_API_KEY=your_gemini_api_key    # Default: transcript generation
OPENAI_API_KEY=your_openai_api_key    # Default: TTS
Optional (for other TTS models)
ELEVENLABS_API_KEY=your_elevenlabs_api_key

Supported TTS Models: | Model | Quality | API Key Required | |:---|:---|:---| | OpenAI (default) | Good | Yes | | ElevenLabs | Great (customizable) | Yes | | Google (gemini/geminimulti) | Best (English only) | Yes | | Edge | Basic | No |

Recommended Setup:

Usage

Basic Examples

#### From URLs (websites, articles)

python -m podcastfy.client --url https://example.com/article1 --url https://example.com/article2

#### From YouTube Video

python -m podcastfy.client --url https://youtube.com/watch?v=xxx

#### From PDF

python -m podcastfy.client --file path/to/document.pdf

#### From Images

python -m podcastfy.client --image path/to/image1.jpg --image path/to/image2.png

#### From Text

python -m podcastfy.client --text "Your raw text content here"

#### Longform (30+ minutes)

python -m podcastfy.client --url https://example.com/article --longform

#### Only Generate Transcript (no audio)

python -m podcastfy.client --url https://example.com/article --transcript-only

Python API Usage

from podcastfy.client import generate_podcast
From URLs
audio_file = generate_podcast(
    urls=["https://example.com/article1", "https://example.com/article2"]
)
From local files
audio_file = generate_podcast(
    file="path/to/urls.txt"
)
From images
audio_file = generate_podcast(
    image=["path/to/image1.jpg", "path/to/image2.png"]
)
Longform
audio_file = generate_podcast(
    urls=["https://example.com/article"],
    longform=True
)

Advanced Options

# Custom conversation config
python -m podcastfy.client --url https://example.com/article --conversation-config config.yaml
Use specific TTS model
python -m podcastfy.client --url https://example.com/article --tts-model elevenlabs
Use local LLM for privacy
python -m podcastfy.client --url https://example.com/article --transcript-only --local
Output to specific directory
python -m podcastfy.client --url https://example.com/article --output-dir ./my-podcasts

Integration with OpenClaw

Method 1: Direct CLI Execution

Use OpenClaw's exec tool:

python -m podcastfy.client --url https://example.com/article

Method 2: Python Script Generation

Generate a Python script and run it:

# Script will be created by the skill
from podcastfy.client import generate_podcast
audio = generate_podcast(
    urls=[""],
    output_dir="./podcasts"
)

Use Cases for Dental/Medical Content

Transform Articles into Podcasts

# Convert dental research articles to audio
python -m podcastfy.client --url https://www.dental.com/research/article

Convert Course Materials

# Turn course PDFs into listenable content
python -m podcastfy.client --file ./dental-course/lecture1.pdf

Create Patient Education

# Transform patient info sheets to audio
python -m podcastfy.client --text "Dental implant procedure explained..."

Summarize YouTube Dental Videos

# Turn dental education videos into podcasts
python -m podcastfy.client --url https://youtube.com/watch?v=dental-education-video

Output

Troubleshooting

"ffmpeg not found"

# Install ffmpeg
brew install ffmpeg  # macOS
sudo apt install ffmpeg  # Linux

"API key not found"

Make sure your .env file is in the correct directory and API keys are set.

"Multi-speaker voices not available"

Google's multi-speaker TTS requires allowlisting. Use OpenAI or ElevenLabs instead.

Resources

Note: Requires Python 3.11+ and FFmpeg.

Basic Examples

#### From URLs (websites, articles)

python -m podcastfy.client --url https://example.com/article1 --url https://example.com/article2

#### From YouTube Video

python -m podcastfy.client --url https://youtube.com/watch?v=xxx

#### From PDF

python -m podcastfy.client --file path/to/document.pdf

#### From Images

python -m podcastfy.client --image path/to/image1.jpg --image path/to/image2.png

#### From Text

python -m podcastfy.client --text "Your raw text content here"

#### Longform (30+ minutes)

python -m podcastfy.client --url https://example.com/article --longform

#### Only Generate Transcript (no audio)

python -m podcastfy.client --url https://example.com/article --transcript-only

Python API Usage

from podcastfy.client import generate_podcast
From URLs
audio_file = generate_podcast(
    urls=["https://example.com/article1", "https://example.com/article2"]
)
From local files
audio_file = generate_podcast(
    file="path/to/urls.txt"
)
From images
audio_file = generate_podcast(
    image=["path/to/image1.jpg", "path/to/image2.png"]
)
Longform
audio_file = generate_podcast(
    urls=["https://example.com/article"],
    longform=True
)

Advanced Options

# Custom conversation config
python -m podcastfy.client --url https://example.com/article --conversation-config config.yaml
Use specific TTS model
python -m podcastfy.client --url https://example.com/article --tts-model elevenlabs
Use local LLM for privacy
python -m podcastfy.client --url https://example.com/article --transcript-only --local
Output to specific directory
python -m podcastfy.client --url https://example.com/article --output-dir ./my-podcasts

1. Install Podcastfy

pip install podcastfy

2. Install FFmpeg (required for audio processing)

# macOS
brew install ffmpeg
Ubuntu/Debian
sudo apt install ffmpeg

3. Configure API Keys

Create a .env file with your API keys:

# Required: At least one LLM + one TTS
GEMINI_API_KEY=your_gemini_api_key    # Default: transcript generation
OPENAI_API_KEY=your_openai_api_key    # Default: TTS
Optional (for other TTS models)
ELEVENLABS_API_KEY=your_elevenlabs_api_key

Supported TTS Models: | Model | Quality | API Key Required | |:---|:---|:---| | OpenAI (default) | Good | Yes | | ElevenLabs | Great (customizable) | Yes | | Google (gemini/geminimulti) | Best (English only) | Yes | | Edge | Basic | No |

Recommended Setup:

"ffmpeg not found"

# Install ffmpeg
brew install ffmpeg  # macOS
sudo apt install ffmpeg  # Linux

"API key not found"

Make sure your .env file is in the correct directory and API keys are set.

"Multi-speaker voices not available"

Google's multi-speaker TTS requires allowlisting. Use OpenAI or ElevenLabs instead.