StarZero All shows Try the podcast clipper →

Oct 1, 2024 · 38m · startup-ideas

Create an AI Voice Character app only using AI in less than 40 min (Cursor, Next.js, Daily Bots)

Eric Min Jun Kim · 25m spoken Greg Isenberg · 6m spoken Weatherman Bot · 15s spoken Minji (AI VTuber) · 13s spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

Greg Isenberg and developer Eric demonstrate how to build, customize, and deploy an AI-powered voice character application in under forty minutes using DailyBots, Next.js, and Cursor. Throughout the build, they explore the technical architecture, function calling mechanics, and massive business opportunities surrounding viral AI personas.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Greg holds 21.3% of the talking time here. How this is scored →

Greg as informed peer 2.6 Guest teaching 4.4 Guest disagreement 0.1 Greg pushing back 0.1
05100:0010:0020:0030:000:44–3:02 · Greg as informed peer 3/10 Welcoming Eric and Overview of Building Voice Agents Greg sets an encouraging stage, framing AI voice characters as the next big wave of consumer apps. Eric enthusiastically outlines what viewers will take away from the practical coding tutorial.3:05–6:04 · Greg as informed peer 1/10 Setting Up Cursor, DailyBots, and Local Environment Eric shares his screen and walks through Cursor, DailyBots, and environment configuration. Greg purely observes while Eric explains how DailyBots simplifies developer orchestration.6:06–12:56 · Greg as informed peer 2/10 Sponsor Break: Greg's Curated Idea Bank Following an ad read, Eric demonstrates the default bot and explains the full STT, LLM, and TTS audio pipeline. Greg asks a basic clarifying question about adding custom characters.12:57–24:31 · Greg as informed peer 6/10 The Business and Viral Potential of Character Apps Greg shares business insight on engineering viral loops and TikTok distribution for character apps. Eric complements this by explaining the technical mechanics of function calling in Next.js.24:33–28:52 · Greg as informed peer 2/10 Sponsor Break: BoringMarketing.com SEO Services After a sponsor break, Eric runs the live weatherman test on his phone, prompting Eric to advise against relying on one-shot AI prompt generation over iterative development.28:54–32:14 · Greg as informed peer 1/10 One-Command Deployment to the Web via Vercel Eric shows how simple deployment is using the Vercel CLI to take the Next.js app live in a single command, with Greg reacting to the ease of modern hosting workflows.32:15–36:33 · Greg as informed peer 3/10 Demoing an Advanced AI VTuber FaceTime App Eric demos his advanced AI VTuber FaceTime app Minji, explaining his shift to PipeCat for deeper developer control. Greg concludes by praising solo builders leveraging AI tools.0:44–3:02 · Guest teaching 2/10 Welcoming Eric and Overview of Building Voice Agents Greg sets an encouraging stage, framing AI voice characters as the next big wave of consumer apps. Eric enthusiastically outlines what viewers will take away from the practical coding tutorial.3:05–6:04 · Guest teaching 4/10 Setting Up Cursor, DailyBots, and Local Environment Eric shares his screen and walks through Cursor, DailyBots, and environment configuration. Greg purely observes while Eric explains how DailyBots simplifies developer orchestration.6:06–12:56 · Guest teaching 6/10 Sponsor Break: Greg's Curated Idea Bank Following an ad read, Eric demonstrates the default bot and explains the full STT, LLM, and TTS audio pipeline. Greg asks a basic clarifying question about adding custom characters.12:57–24:31 · Guest teaching 5/10 The Business and Viral Potential of Character Apps Greg shares business insight on engineering viral loops and TikTok distribution for character apps. Eric complements this by explaining the technical mechanics of function calling in Next.js.24:33–28:52 · Guest teaching 4/10 Sponsor Break: BoringMarketing.com SEO Services After a sponsor break, Eric runs the live weatherman test on his phone, prompting Eric to advise against relying on one-shot AI prompt generation over iterative development.28:54–32:14 · Guest teaching 4/10 One-Command Deployment to the Web via Vercel Eric shows how simple deployment is using the Vercel CLI to take the Next.js app live in a single command, with Greg reacting to the ease of modern hosting workflows.32:15–36:33 · Guest teaching 6/10 Demoing an Advanced AI VTuber FaceTime App Eric demos his advanced AI VTuber FaceTime app Minji, explaining his shift to PipeCat for deeper developer control. Greg concludes by praising solo builders leveraging AI tools.0:44–3:02 · Guest disagreement 0/10 Welcoming Eric and Overview of Building Voice Agents Greg sets an encouraging stage, framing AI voice characters as the next big wave of consumer apps. Eric enthusiastically outlines what viewers will take away from the practical coding tutorial.3:05–6:04 · Guest disagreement 0/10 Setting Up Cursor, DailyBots, and Local Environment Eric shares his screen and walks through Cursor, DailyBots, and environment configuration. Greg purely observes while Eric explains how DailyBots simplifies developer orchestration.6:06–12:56 · Guest disagreement 0/10 Sponsor Break: Greg's Curated Idea Bank Following an ad read, Eric demonstrates the default bot and explains the full STT, LLM, and TTS audio pipeline. Greg asks a basic clarifying question about adding custom characters.12:57–24:31 · Guest disagreement 1/10 The Business and Viral Potential of Character Apps Greg shares business insight on engineering viral loops and TikTok distribution for character apps. Eric complements this by explaining the technical mechanics of function calling in Next.js.24:33–28:52 · Guest disagreement 0/10 Sponsor Break: BoringMarketing.com SEO Services After a sponsor break, Eric runs the live weatherman test on his phone, prompting Eric to advise against relying on one-shot AI prompt generation over iterative development.28:54–32:14 · Guest disagreement 0/10 One-Command Deployment to the Web via Vercel Eric shows how simple deployment is using the Vercel CLI to take the Next.js app live in a single command, with Greg reacting to the ease of modern hosting workflows.32:15–36:33 · Guest disagreement 0/10 Demoing an Advanced AI VTuber FaceTime App Eric demos his advanced AI VTuber FaceTime app Minji, explaining his shift to PipeCat for deeper developer control. Greg concludes by praising solo builders leveraging AI tools.0:44–3:02 · Greg pushing back 0/10 Welcoming Eric and Overview of Building Voice Agents Greg sets an encouraging stage, framing AI voice characters as the next big wave of consumer apps. Eric enthusiastically outlines what viewers will take away from the practical coding tutorial.3:05–6:04 · Greg pushing back 0/10 Setting Up Cursor, DailyBots, and Local Environment Eric shares his screen and walks through Cursor, DailyBots, and environment configuration. Greg purely observes while Eric explains how DailyBots simplifies developer orchestration.6:06–12:56 · Greg pushing back 0/10 Sponsor Break: Greg's Curated Idea Bank Following an ad read, Eric demonstrates the default bot and explains the full STT, LLM, and TTS audio pipeline. Greg asks a basic clarifying question about adding custom characters.12:57–24:31 · Greg pushing back 1/10 The Business and Viral Potential of Character Apps Greg shares business insight on engineering viral loops and TikTok distribution for character apps. Eric complements this by explaining the technical mechanics of function calling in Next.js.24:33–28:52 · Greg pushing back 0/10 Sponsor Break: BoringMarketing.com SEO Services After a sponsor break, Eric runs the live weatherman test on his phone, prompting Eric to advise against relying on one-shot AI prompt generation over iterative development.28:54–32:14 · Greg pushing back 0/10 One-Command Deployment to the Web via Vercel Eric shows how simple deployment is using the Vercel CLI to take the Next.js app live in a single command, with Greg reacting to the ease of modern hosting workflows.32:15–36:33 · Greg pushing back 0/10 Demoing an Advanced AI VTuber FaceTime App Eric demos his advanced AI VTuber FaceTime app Minji, explaining his shift to PipeCat for deeper developer control. Greg concludes by praising solo builders leveraging AI tools.

speaking balance: gold is Greg, purple is the guest (3 minute bins)

0:00 · Greg 51.3% · guest 48.7%0:00 · Greg 51.3% · guest 48.7%3:00 · Greg 1.3% · guest 98.7%3:00 · Greg 1.3% · guest 98.7%6:00 · Greg 33.1% · guest 66.9%6:00 · Greg 33.1% · guest 66.9%9:00 · Greg 0% · guest 100%9:00 · Greg 0% · guest 100%12:00 · Greg 24.4% · guest 75.6%12:00 · Greg 24.4% · guest 75.6%15:00 · Greg 0% · guest 100%15:00 · Greg 0% · guest 100%18:00 · Greg 0% · guest 100%18:00 · Greg 0% · guest 100%21:00 · Greg 42.9% · guest 57.1%21:00 · Greg 42.9% · guest 57.1%24:00 · Greg 42.4% · guest 57.6%24:00 · Greg 42.4% · guest 57.6%27:00 · Greg 19.5% · guest 80.5%27:00 · Greg 19.5% · guest 80.5%30:00 · Greg 14.3% · guest 85.7%30:00 · Greg 14.3% · guest 85.7%33:00 · Greg 0.9% · guest 99.1%33:00 · Greg 0.9% · guest 99.1%36:00 · Greg 63.5% · guest 36.5%36:00 · Greg 63.5% · guest 36.5%
Sharpest disagreement ▶ 15:30 Debunking AI black box myths around function calling

Eric dispels common misconceptions by firmly clarifying that LLMs cannot perform actions on their own and only output structured text that surrounding code executes.

 Hardest push from Greg ▶ 12:16 Questioning character customization limits

Greg probes whether users are restricted to preset characters, prompting Eric to prove how directly the config file maps to new character additions.

 Biggest teaching moment ▶ 9:30 Full architectural explanation of real-time voice agents

Eric gives a clear, step-by-step masterclass on how audio inputs pass from speech-to-text into an LLM and then through text-to-speech back to the browser.

 Greg holds their own ▶ 22:55 Strategic framework for designing viral AI characters

Greg demonstrates sharp product strategy by explaining how building deliberate quirks and constraints into an AI character reverse-engineers viral social media clips.
the scores for every segment, with the reasoning behind each
ChapterTopicGreg as informed peerGuest teachingGuest disagreementGreg pushing backWhy
Welcoming Eric and Overview of Building Voice Agents 3200 Greg sets an encouraging stage, framing AI voice characters as the next big wave of consumer apps. Eric enthusiastically outlines what viewers will take away from the practical coding tutorial.
Setting Up Cursor, DailyBots, and Local Environment 1400 Eric shares his screen and walks through Cursor, DailyBots, and environment configuration. Greg purely observes while Eric explains how DailyBots simplifies developer orchestration.
Sponsor Break: Greg's Curated Idea Bank 2600 Following an ad read, Eric demonstrates the default bot and explains the full STT, LLM, and TTS audio pipeline. Greg asks a basic clarifying question about adding custom characters.
The Business and Viral Potential of Character Apps 6511 Greg shares business insight on engineering viral loops and TikTok distribution for character apps. Eric complements this by explaining the technical mechanics of function calling in Next.js.
Sponsor Break: BoringMarketing.com SEO Services 2400 After a sponsor break, Eric runs the live weatherman test on his phone, prompting Eric to advise against relying on one-shot AI prompt generation over iterative development.
One-Command Deployment to the Web via Vercel 1400 Eric shows how simple deployment is using the Vercel CLI to take the Next.js app live in a single command, with Greg reacting to the ease of modern hosting workflows.
Demoing an Advanced AI VTuber FaceTime App 3600 Eric demos his advanced AI VTuber FaceTime app Minji, explaining his shift to PipeCat for deeper developer control. Greg concludes by praising solo builders leveraging AI tools.

Statements from this episode (9)

Opinion
Kim: Voice interaction with ChatGPT is a 10x improvement over text
“Like, you experience ChatGPT for the first time, and you're like, holy shit, this is cool. But then you talk to ChatGPT, and it's almost like this 10 X improvement in experience.”
Eric Min Jun Kim Oct 1, 2024 ▶ 1:33
Opinion
Kim: DailyBots simplifies development by abstracting backend LLM orchestration
“What DailyBots, when they came out with this, I was like, oh, This is going to be kind of game changing because now you don't even have to handle all the LLM calls and stuff with pipe cat daily is doing that themselves.”
Eric Min Jun Kim Oct 1, 2024 ▶ 4:16
Insight
Kim: The essential first step in building AI apps is inspecting the raw prompt
“Whenever I'm working with AI stuff, I'm always like, the first thing that I think to myself is like, show me the prompt. I want to know what exactly is going into the LLM.”
Eric Min Jun Kim Oct 1, 2024 ▶ 10:07
Insight
Isenberg: AI characters are the new apps, creating a massive opportunity
“I think that if you think of characters as the new apps, Then there's a huge opportunity, and not many people are talking about this, around just ideating on characters, figuring out what's gonna work, and launching them.”
Greg Isenberg Oct 1, 2024 ▶ 13:00
Assertion Supported
Kim: Nikita Bier changed his mind on AI characters after seeing retention
“And what Nikita said in that tweet was that initially he thought like AI character chat apps were kind of a fad, but after seeing like a bunch of the retention graphs, he's like, okay, this is not going away.”
Eric Min Jun Kim Oct 1, 2024 ▶ 13:55
Prediction Not checkable as stated
Isenberg: AI characters will become the primary way we interact with content
“And I think it's B to B and B to C. I think it's just going to be the way that we're going to interact with content. And this is just a new form of content.”
Greg Isenberg Oct 1, 2024 ▶ 14:30
Insight
Isenberg: Add constraints to AI characters to engineer viral distribution
“As you're designing your character and thinking about you know, the ultimate experience that you want to create. I think it is important to add constraints and try to think about what's the type of thing that will get you free distribution.”
Greg Isenberg Oct 1, 2024 ▶ 23:25
Assertion Supported
Kim: Top VTubers make millions annually from subscriptions and merchandise
“They have, like, Millions of followers. They're making millions of dollars a year from subscriptions, from selling merch, et cetera.”
Eric Min Jun Kim Oct 1, 2024 ▶ 33:30
Insight
Isenberg: Building AI Character Apps Previously Required VC and Large Teams
“A few years ago, well, first of all, it wasn't, this wasn't even possible a few years ago, but you would need a large team to do that, raise venture capital.”
Greg Isenberg Oct 1, 2024 ▶ 36:56
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 300 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more