Overview

Podcast Generator is a powerful tool for creating podcast audio files using Text-to-Speech (TTS) services. This tool can generate a podcast with alternating voices by providing a script, making it ideal for dialogue-based content, interviews, or storytelling.

Powered by OpenAI-based TTS services, Podcast Generator simplifies the production of high-quality audio content. Currently this tool supoorts following TTS services:

• OpenAI TTS (and other OpenAI-compatible TTS services)
• Azure OpenAI TTS

You can specify the TTS models, voices, and instructions used for each host in the podcast, allowing for a personalized touch to your audio content.

Configuration

To start using Podcast Generator, follow these steps:

1. Get an API Key for your TTS Service
   Depending on the TTS service you want to use (for example, OpenAI), create a new API Key, and ensure that your account is properly set up with sufficient access.
2. Install the Podcast Generator Tool
   Open the Plugin Marketplace, search for the Podcast Generator tool, and install it to integrate it with your application.
3. Authorize Podcast Generator
   Go to Plugins > Podcast Generator > Authorize.
   Enter your API Key to activate the tool. Here's a minimal example of how to set up the tool:
   
   

You can authorize multiple TTS services or models by repeating the authorization process for each service or model you want to use.
It's recommended to name each authorization entry by Authorization Name field, so you can easily identify which TTS service or model to use later.

Note: If you select OpenAI TTS and leave the Model Name field empty, it's defaulted to . If you select Azure OpenAI TTS, the API Base URL and Deployment Name fields are required.

Tool Feature

The Podcast Audio Generator action converts your text script into a podcast audio file using two alternating voices.

1. Input the Podcast Script
   Provide a script containing alternating lines for two hosts, separated by newline characters. Each line corresponds to what one of the hosts will say.
2. Select Host Voices
   Choose a voice for Host 1 and Host 2 from the available options. These voices will be used to narrate the respective lines in the script.
   Optionally, you can also specify the Instructions for each host to guide the tone or style of their narration.
3. Select Output Format
   By default, the output format is monaural WAV audio. You can select or in the Channel Mode field, and specify or as the output format in the Output Format field.
   If you select mode, Host 1's voice is positioned slightly to the left, while Host 2's voice is positioned slightly to the right.
4. Generate the Podcast Audio
   Once the script and voices are set, run the action to create your podcast audio file. The result will combine the two voices, alternating as per the script.

Usage

Podcast Generator can seamlessly integrate Chatflow / Workflow Apps and Agent Apps.

Chatflow / Workflow Apps

1. Add the Podcast Generator node to your Chatflow or Workflow pipeline.
2. Input the podcast script in the node configuration.
3. Run the pipeline to generate and deliver the podcast audio as part of the workflow.

Agent Apps

1. Add the Podcast Generator tool to your Agent application.
2. Send a podcast script via the chat interface.
3. The tool processes the script and returns the generated audio file to the user.

Use Cases

• Podcast Creation: Quickly produce podcasts for storytelling, interviews, or educational content.
• Dialogue Simulations: Generate podcasts with alternating voices to mimic conversations or debates.
• Marketing Content: Create engaging audio advertisements or announcements.
• Content Accessibility: Provide audio versions of written content for visually impaired audiences.

The Podcast Generator makes podcast production efficient and accessible, enabling creators to focus on crafting compelling scripts while automating the audio generation process.