Note: When I refer to "programming" in this post, I'm talking about writing code (like Python, JavaScript, etc.), not like in the context broadcast/network programming.
A year ago, I left my corporate job to pursue podcast post-production full-time. As someone with a background in programming, I wasn't sure how my skills would translate to this new field. I often felt intimidated by other producers who seemed to have incredible creative chops that I lacked.
But over time, I realized that my programming background could be a unique strength in podcast production. By focusing on automating workflows and building internal tools, I found ways to be creative that played to my strengths. Today, I want to share some insights I've gained and tools I've found useful in my journey as a podcast producer.
The Challenge: Balancing Repetitive Tasks and Creativity
As podcast producers, we often find ourselves bogged down by numerous repetitive, manual tasks. These can include editing audio, creating show notes, repurposing content for social media, and more. The time spent on these routine activities often leaves little room for the creative aspects of podcast production.
The key to overcoming this challenge lies in automation. By automating repetitive tasks, we can free up more time to focus on the creative elements that truly make a podcast shine. If you have programming and coding skills, you're in a unique position to build tools and workflows that can significantly streamline your production process.
In this post, I'm going to share some of the tools and approaches I've developed or adopted to automate various aspects of podcast production. These solutions have allowed me to spend more time on the creative tasks that I enjoy and that add the most value to the podcasts I produce.
Most of the solutions I recommend are open-source and free. As a strong believer in open-source software, I prefer these solutions for their accessibility, customizability, and community-driven development. But that is a personal choice and I have no affliation with any of these tools.
Open-Source Tools and Automation in Podcast Production
Here are some open-source tools and approaches I use in my work:
1. FFMPEG: Versatile Media Processing
FFMPEG is a command-line tool for handling audio and video tasks. It can convert file formats, extract audio from video, trim files, and perform many other media operations. I use FFMPEG to standardize podcast audio formats and add intros and outros to episodes automatically. It serves as the foundation for many of my automation scripts, handling tasks that would otherwise require multiple specialized programs.
Think of this as a Swiss Army knife for media processing. IMO, it is the best open source software ever created. If you need to learn and master only one programming tool for media processing, I would recommend this. Intimidating at first but totally worth it. It can do everything encoding, decoding, transcoding, trimming and such.
2. ClipsAI: Efficient Video Repurposing
ClipsAI is a tool that automatically crops videos to different aspect ratios. It's particularly useful for adapting landscape video recordings to formats suitable for various social media platforms. The tool uses AI to identify the speaker parts of a video when cropping. I use ClipsAI to create Instagram and TikTok clips from full-length podcast videos, which significantly reduces the time spent on manual video editing.
It is basically Fffmpeg, Python and OpenCV wrapped together underneath. It's open source. If you don't want to build your own software, this is just nice plug and play.
3. Large Language Models: Automated Content Writing
Large language models like LLaMA are AI tools that can generate written content based on prompts. These models can assist in creating show notes, episode descriptions, and other text-based content for podcasts. The quality of the output depends on the effectiveness of the prompts used. In my workflow, I use these models to generate first drafts of show notes, which I then review and edit as needed.
You can also use OpenAI APIs, but I just like open source LLM tools because I don't have to depend on external vendors.
4. Custom Ad Insertion Script: Streamlined Episode Assembly
This is not a tool but rather an example. One of the first automation projects I tackled was creating a script to assemble podcast episodes. Many of our shows follow a standard structure: intro, content, ad spots, and outro. Manually putting these pieces together for each episode was tedious and error-prone. My custom script now handles this process automatically, ensuring consistency and saving a ton of time, especially when we're producing multiple episodes a week. It's a simple tool, but it's had a big impact on our workflow.
Just feed in a bunch of files, and tell where you want the midrolls to go. And bam we have the final files. Probably the highest leverage tool that I built. Built with a bit of programming with ffmpeg and python. Ugly mess of a code, but hey it works!
5. Nvidia Audio SDKs: Audio Enhancement
Nvidia provides free audio software development kits (SDKs) that offer tools for noise removal and audio enhancement. These SDKs are designed to work with Nvidia GPUs and can achieve results comparable to some commercial audio cleanup software. I use these tools to improve the audio quality of podcast recordings, particularly for episodes recorded in suboptimal acoustic environments.
The software itself is free. Of course, you need to have a PC with GPU to run it.
6. Pipedream: Workflow Automation
Pipedream is a platform for creating automated workflows. It's like Zapier, but built for developers and programmers. A no-code automation tool with more flexibility. As I developed more tools and scripts, I needed a way to tie everything together. Pipedream has been the perfect solution for this.
I've set up processes that handle everything. For example, I don't manually upload to YouTube, Spotify and Apple anymore. Setup a pipedream automation, drag in files and show notes and 10 mins later its published to all platform. While there was definitely a learning curve, the flexibility Pipedream offers has allowed me to create a nearly end-to-end automated production pipeline, freeing up more time for the creative aspects of podcast production.
Practical Tips for Podcast Producers
Whether you're new to podcast production or looking to optimize your workflow, here are some tips based on my experience:
- Learn basic scripting: Even a little knowledge of a language like Python can help you automate repetitive tasks.
- Explore command-line tools: Tools like FFMPEG can be powerful additions to your toolkit.
- Try out AI tools: Experiment with AI for tasks like writing show notes, but remember to review and refine the output. Reliability is not 100% but can be close to 100% with effort.
- Look for repetitive tasks: These are often good candidates for automation.
- Consider open-source alternatives: Before investing in expensive software, check if there are open-source options that could work for you. In my experience, most of them are wrapper around open source tools + LLMs. Especially with recent boom around AI. I am not dunking on them. If you have no idea how to program, these paid tools are super useful. But if you are willing to put in a little effort, you will save ton of money and also they are much more flexible.
Conclusion: Finding Your Unique Approach
When I started in podcast production, I felt out of place because I didn't have the same "creative" skills as many other producers. However, I've learned that there are many ways to be "creative" in this field. For me, creativity comes through in how I build workflows and tools to support the podcasters/podcasts I work with.
If you're coming from a coding background like me, don't underestimate the value of your skills in podcast production. You can create unique solutions that make your work more efficient and effective. This took me months to realize.
At the same time, it's important to remember that these tools and automations are just that - tools. They're here to support the creative process, not replace it. The goal is to free up more time for the aspects of podcast production that require human creativity and insight.
I'm curious to hear about your experiences. How have you applied your unique skills to podcast production? What challenges are you still facing in your workflow? Any favorite tools (preferable open source :D) tools? I am always on hunt for new tools.
