podfeed-sdk 0.6.0

A Python SDK for the PodFeed API

PodFeed SDK

A Python SDK for the PodFeed API, enabling developers to generate high-quality Podcast-style audio content from various input sources using AI.

Features

• Multiple Input Types: Support for text, URLs, files, topics, and bring-your-own-script
• Combined Sources: Combine multiple URLs, text blocks, and files in a single request (up to 10 sources)
• Pre-add to Feed: Optionally add generated audio to a podcast feed automatically via feed_id
• Feed Management: List feeds, create feeds, add existing audios to feeds
• Audio Generation Modes: Monologue (single voice) and dialogue (two voices) modes
• Voice Customization: Multiple voice options for different languages. Some voices support custom instructions
• Script Customization: Adjustable complexity levels, lengths, and emphasis

Installation

pip install podfeed-sdk

Authentication

The SDK supports two ways to provide your API key:

1. Environment Variable:

   export PODFEED_API_KEY="your-api-key-here"
   

2. Direct initialization:

   from podfeed import PodfeedClient
   
   client = PodfeedClient(api_key="your-api-key-here")
   

Quick Start

Get an API key

Go to https://podfeed.ai, log in, and get an API key.

Generate Audio (from Website)

from podfeed import (
    PodfeedClient,
    PodfeedError,
    AudioGenerationRequest,
    InputContent,
    VoiceConfig,
    ContentConfig,
)

# Initialize client (uses PODFEED_API_KEY env var)
client = PodfeedClient()

# Example URL
website_url = "https://podfeed.ai/faq"

# Generate audio from website
result = client.generate_audio(
    request=AudioGenerationRequest(
        input_type="url",
        mode="dialogue",
        input_content=InputContent(url=website_url),
        voice_config=VoiceConfig(
            host_voice="google-male-puck", cohost_voice="google-female-leda"
        ),
        content_config=ContentConfig(
            level="intermediate",
            length="medium",
            language="en-US",
        ),
    )
)

task_id = result["task_id"]
print(f"Task created: {task_id}")

# Wait for completion
final = client.wait_for_completion(task_id)
print(f"Audio generated: {final['result']['audio_url']}")

Combined Sources (Text + URL + File)

Combine multiple inputs in one request:

from podfeed import SourceItem

result = client.generate_audio(
    request=AudioGenerationRequest(
        content_type="sources",
        sources_content=[
            SourceItem(input_type="text", content="Introduction to AI..."),
            SourceItem(input_type="url", content="https://example.com/article"),
        ],
        mode="dialogue",
        voice_config=VoiceConfig(host_voice="...", cohost_voice="..."),
        content_config=ContentConfig(language="en-US"),
    )
)

Pre-add to Feed

Add generated audio to a feed automatically:

result = client.generate_audio(
    request=AudioGenerationRequest(
        input_type="topic",
        input_content=InputContent(topic="Quantum computing"),
        mode="dialogue",
        voice_config=VoiceConfig(host_voice="...", cohost_voice="..."),
        content_config=ContentConfig(language="en-US"),
        feed_id="your-feed-id",  # Audio added to feed when ready
    )
)

List Available Voices

from podfeed import PodfeedClient

api_key = os.getenv("PODFEED_API_KEY")
if not api_key:
    print("Error: PODFEED_API_KEY environment variable not set")
    return 1

client = PodfeedClient(api_key=api_key)

voices_config = client.list_available_voices()
print(voices_config)

Usage Examples

See the example in the examples repository: https://github.com/smh-labs/podfeed-sdk-samples

Error Handling

from podfeed_sdk import PodfeedClient, PodfeedError, PodfeedAuthError, PodfeedAPIError

try:
    client = PodfeedClient()
    response = client.generate_audio(
        input_type="text",
        text_content="Sample text"
    )
except PodFeedAuthError as e:
    print(f"Authentication error: {e}")
except PodFeedAPIError as e:
    print(f"API error: {e.message} (Status: {e.status_code})")
except PodFeedError as e:
    print(f"General error: {e}")

API Reference

We'll be publishing an API reference and documentation soon.

Requirements

• Python 3.7+
• requests >= 2.25.0

Rate Limits

Details coming soon.

Support

For API support, email [e-mail].