KludgeHack

 What is Feed to Podcast?

Feed to Podcast is a simple application built on C# and .NET 3, making use of the new sub-system on the framework: Windows Presentation Foundation (WPF).

As my first attempt to build a somehow useful and simple to use application leveraging this new technologies, this application takes a RSS address from the user and converts its content to sound files (podcasts) compatible with the most popular devices (Zune, iPod, etc).

How it Works?

With the new framework versions, the Speech API was improved, so as the new text-to-speech (TTS) voice in Windows Vista: Microsoft Anna. With this improvement it’s really easy to build an application around this, and make applications “sound like humans”. More and more, the voices will only get more natural.

The code around the Speech API and the WM Encoder – required to encode wave files to WMA – its based on the article written by Paul-Valentin Borza. Some tweaks have been made to allow more flexibility in the configuration and asynchronous work.

How can I download the Application?

First of all, you need to have these prerequisites:

• Microsoft .NET Framework 3.0 Redistributable Package (already installed on Windows Vista)
• Windows Media Encoder 9 Series to encode Wave files to Windows Media Audio (WMA). The WMA format is compatible with Zune, iPod, WMP Mobile, WMP, etc.

After this, you can download the application binaries if you want to test as it is, or download the source code, if you want to develop something even better:

•  Feed to Podcast – Binaries
•  Feed to Podcast – Source Code

The code is still very *alpha beta*, as this is more a proof-of-concept.

How to use it?

Just open the application and enter the RSS address in the textbox. If you want to preview the feed, you can click in ‘Preview’ to show a list of posts. Click ‘process’ to create the audio files in the folder chosen in the configuration (default: C:\Podcast).

Known issues:

• Only RSS support (no ATOM)
• Only WMA support. Mp3 support would be even better.
• The current method to clean HTML posts back to Plain text is not very good, and some podcasts end up with some garbage audio inside.