Earkind
Earkind is an AI tool that generates podcasts with a focus on Artificial Intelligence news, research paper dives, and joke attempts. The to...
Last verified:
What is Earkind?
Earkind is an AI-generated podcast platform that creates entertaining daily podcasts about Artificial Intelligence news, research paper deep dives, and joke attempts. The flagship show
Earkind pricing
Pricing model: Free
Earkind is completely free to listen to. The podcast is available on Spotify, Amazon Music, and Apple Podcasts with no subscription fee or paid tiers required. Users can access all AI-generated podcasts, enhanced podcast editing features, and automated podcast creation without any cost. The creator is open to feedback and has not announced any paid plans.
Earkind pros
- AI-generated podcasts that don't bore listeners
- Daily updates on AI news and developments
- Covers arXiv research paper deep dives
- Entertaining humorous commentary on AI topics
- Multiple distinct character voices with unique personalities
- Available on Spotify, Amazon Music, and Apple Podcasts
- Free to listen with no subscription required
- Open to user feedback and improvement
- Plans to make code publicly available
- Azure's top-notch neural text-to-speech technology
- Automated podcast description with timestamps
- Professional audio editing with jingles and sound effects
- Combines news with research paper analysis
- Non-serious over-the-top presentation style
- Diverse perspectives from multiple hosts and collaborators
Earkind cons
- Very rough early-stage product quality
- Limited to AI-related topics only
- Only one show available (GPT Reviews)
- No personalized content generation for users
- Characters can be annoying (overly hyped tech bro)
- No user control over episode length or topics
- Creator handles all news selection manually
- Noise in extracted PDF text from arXiv
Frequently asked questions about Earkind
What is Earkind?
Earkind is an AI-generated podcast platform that creates entertaining daily podcasts focusing on Artificial Intelligence news, research paper dives from arXiv, and joke attempts. It combines language models with neural expressive text-to-speech and programmatic audio editing to create full podcast episodes.
What is the GPT Reviews show?
GPT Reviews is Earkind's flagship daily show that comments on Artificial Intelligence news, includes joke attempts, and dives into research papers. It features host Giovani Pete Tizzano (an overly hyped AI follower) and collaborators Robert (an often unimpressed analyst), Olivia (an overly online internet surfer), and Belinda (an all-around witty research expert).
Where can I listen to Earkind podcasts?
Earkind podcasts are available on popular podcast listening apps including Spotify, Amazon Music, and Apple Podcasts, making the content accessible to a wide audience across all major platforms.
How does Earkind create podcasts?
The process starts with a text file containing selected news and 3 arXiv paper URLs. Titles and abstracts are crawled from arXiv, info is extracted from raw PDF text using chatGPT API. Tuned prompts create scripts for each section (intro, outro, transitions, sponsor, news, papers). Scripts are recorded using Azure's TTS, then edited with Pydub combining jingles, sound effects, and music. Finally, descriptions with timestamps are automatically generated.
Is Earkind free to use?
Yes, Earkind is completely free to listen to. There are no subscription fees, paid tiers, or costs required to access the AI-generated podcasts on Spotify, Amazon Music, or Apple Podcasts.
How often are new episodes released?
GPT Reviews is a daily show, meaning new episodes are released every day covering the latest AI news, announcements, and research from arXiv mixed with fun content.
Can I suggest topics or news for the podcast?
The creator actively welcomes feedback and encourages contact at [email protected]. All feedback is welcome, and the creator is keen to hear thoughts on whether people would be interested in this type of personalized audio content.
Will the Earkind code be publicly available?
Yes, the creator plans to make the code public and publish a more in-depth explanation on the design and thinking behind the system if there is interest from the audience.
What AI technologies does Earkind use?
Earkind uses language models (LMs) for content generation, the chatGPT API for extracting information from PDFs, Azure's neural expressive text-to-speech for recording (based on Microsoft research), and Pydub for programmatic audio editing including volume adjustment, section overlaying, and looping.
Who created Earkind?
Earkind was created by Sergi (contact: [email protected]), who developed the pipeline combining LMs with neural TTS and audio editing. The podcast features host Giovani Pete Tizzano and collaborators Robert, Olivia, and Belinda as AI-generated characters with distinct personalities.