What we are thinking about

Anthony Moore, aka (@FatmaNDev), created a compelling Podcast skill for Amazon Alexa. VoiceLabs is proud to showcase him.

What sparked Pod Buddy?

I originally started the project for myself – to play the common podcast I listen to. I created the basic functionalities, like search and start from a specific point (i.e. the 5-minute mark), to emulate the experience of using a podcast app on a phone.

One day, when I was talking to someone at work, he told me it was actually cool and I should think about publishing it publicly it. To release a basic version, I spent hours looking for RSS feeds and aggregating them into the skill. In early January, I published it and built a website to gather feedback. I got a lot more web traffic than I expected, and a Reddit post made the skill popular to the point that people started asking for it in the UK.

General thoughts on building on Alexa Amazon?

Amazon Lambda makes it easier. I decided to use Amazon Lambda Free Tier to publish the skill and see if Amazon would accept it. The nice thing about Lambda is the easy scalability and the ability to test a lot of things locally and push it out to Lambda. Lambda has a great CLI to push things out. Setting up Lambda was very simple: you have an Alexa skill, it automatically builds a trigger for Alexa, you select the type of language you want to write in and you upload the source file.

Previously, I built a trivia app which required hosting elsewhere. Setting up of the database and infrastructure became expensive and cumbersome.

The only difficult part of building Pod Buddy was the podcast functionality from the Alexa SDK. Last October there wasn’t too much documentation, but enough to figure it out. It took time to make sure the structure matches.

Non-Lambda hosted skills need a better IDE. The trivia app required a lot of trial and error pushing. First, I would push to Host, and then on my Amazon Echo try and make sure things worked or heard me. Relatively, testing in Lambda is easier because you’re provided the object that gets passed through the console.

Best practices and thoughts on future of podcasting?

For Building Podcasts. There are a few things to think about when building a podcast. First, make sure to build great basic features like fast forward or rewind. Second, look at features that aren’t available for TuneIn, like search-ability of the podcast. For instance, it’s hard to make a query-request like “Alexa, play the most recent podcast with X Guest on it” or “Alexa, play episode number X.” The main goal of having good search-ability is to help people get to the content they’re looking for. If people wanted the most recent episode they would be going to TuneIn and if not, they would want a great podcast experience. Overall, I believe the goal of home automation is to make someone’s life easier and I know people don’t want the bare minimum. Therefore, If you’re building an Alexa Skill, put a little effort to add the features in that matter. If you’re building a podcast, you can make search-ability even more robust. I’m only using basic regex matching right now, but you can do indexes and create a great experience.

Podcast Tips. A few lessons learned

  1. Certain podcasts can’t stream because it doesn’t have HTTPS or it’s not an MP3 file.
  2. Video podcasts are difficult because of sound quality, although there is demand for them.

Amazon Echo Show. I am curious to see what new things developers will be given in the updated SDK from Amazon. The Fire TV devices do support Alexa, but there is currently a limitation on what can visually be displayed. An idea for podcasting is to start passing video streams to new devices and do a video podcast with the ability to interact with your voice. I’m generally excited to see what other cool experience emerge through visual screens with the seamless voice-interactions.

Interactive Podcasting. I believe the various podcasts help you feel affinity for a wide range of interests. Kevin Smith did a podcast live and he held a Fan Q&A during the show, someone asked: “When you were making the movie Clerks, what did you think about when you did X?” There was also interaction on Twitter for questions from people listening live elsewhere. It would be interesting to see a similar live-to-air interaction on a bigger scale with Alexa at Home and on the new Amazon Echo Show, which can take entertainment to the next level. Also, when Amazon supports live audio, it would be great to be able to tune into a live podcast or a Twitch show and interact with your voice.

It has been interesting to see how podcasting has expanded over the past 10 years. Early on, I would never listen to podcasts, but now I can find a podcast about anything I’m interested in, and it’s a great way to get your voice heard. The podcast ecosystem has been a good forum to hear what people have to say, learn new stuff, and make people feel included in the community.

Thoughts on other platforms?

I took a look at the Google Home, but don’t know if I want to buy another home device. I bought the Amazon Echo right when it came out. Also, when the smaller, Echo Dot came out, I bought one for the kitchen and left the Echo in the office.

I didn’t buy a Google Home because the switching cost is high and once I’m already familiar with one system, it’s hard to adapt to a new one. I would really need to be annoyed with the current OS to consider getting a different device. I recently played around with Alexa Calling, but I’ve been using Google Voice since inception. I think calling is going to be huge and this could be the new landlines — calling people from home and not your cell. It’ll also be interesting to see voice messages versus texting or even more transcriptions from speech-to-text over Alexa or Google Home.

What is next for the Pod Buddy Skill?

Recently, I released Version 2, giving people the ability to shuffle and perform continuous play. Also, I added more direct intents. Most of these changes were based on feedback directly from users.

There are a few things I’d like to add like:

  1. Rewind
  2. Listen in sequential order to a podcast
  3. Resume where you left off for each podcast

Some of the basic functionality was not obvious when I first started out, but I got a lot of organic feedback from the Alexa Skill Store and on the Reddit thread. I’ll sometimes get a request for specific podcasts as well.

I hope people can try out Pod Buddy and send over any feedback.

Last questions, how did you get into programming? And where did “FatmaN Dev” come from?

I was first exposed to computers around the age of 4 or 5 years old and ran DOS, but didn’t really own a computer till high school. I took my first web development course in high school in Java and self-taught myself C during that time. Originally, I was not too sure what to do post high school and just took a job at a call center. Meanwhile, at the call center, I started picking up programming to automate certain tasks and make my job easier.

After a period of time at the company, I met with one of the supervisors. During our conversation, he thought I was really knowledgeable about the company and function, and he asked me if I wanted to do more software engineering. Since then, I became a full-time software engineer and continued to build more. In addition, I recently went back to school, at a local community college, for a couple semesters to learn more about core programming concepts. I wrote the Pod Buddy Skill in Python and I never really picked up Python until I built the skill. I was primarily proficient in Ruby and building Pod Buddy helped me learn Python.

When Alexa first came out, I wasn’t sure if I would like it. However, the more and more I used it, it became great, and I then found myself exploring other smart home devices which connected with Alexa.

The name, “FatmaN,” has a good story, but the main thing is, I’m not really a small guy. When I was in middle school, I decided to embrace my size and to not take myself too seriously about being big. A lot of people from school still recall me for my nickname and it’s kind of stuck with me throughout the years, hence FatmaN Dev.