Live data from Hacker News

Build an AI telephony agent for inbound and outbound calls

github.com

41–50 of 50 posts

Re: Build an AI telephony agent for inbound and outbound calls

#41
post #29
post #27

we tried to build something similar lately for outbound calls (for simple reminders to partners) and faced massive issues using gpt-4o-realtime-audio. Noise detection, turn detection, random telephony issues (we were using Twilio too), prompt not holding together, and more. We dropped the project because it would have resulted in a terrible experience for the person on the other side of the phone. Building these thin…

I've read this comment twice and I genuinely can't understand it. Uh, so your own attempt at a similar project didn't work and was a terrible experience and the fundamentals of the system are specific and still require babysitting. But it's inevitable (???) that it'll get better... and this improvement only MIGHT make things better for people, only some of the time? I'm not alone in being unimpressed by this, right?…

Fair point. But when implemented properly, these agents can reliably handle narrow, production-grade tasks like appointment reminders or smart call routing.

Re: Build an AI telephony agent for inbound and outbound calls

#42
post #32
post #29

Earlier quoted context omitted.

I've read this comment twice and I genuinely can't understand it. Uh, so your own attempt at a similar project didn't work and was a terrible experience and the fundamentals of the system are specific and still require babysitting. But it's inevitable (???) that it'll get better... and this improvement only MIGHT make things better for people, only some of the time? I'm not alone in being unimpressed by this, right?…

what I mean is: building these systems is nontrivial, but if done well it can help. Imagine non being in an endless queue on a phone call when having to do a simple task through a customer center call, or having a phone reminder with more information and less noise than from a written notification. The fact that I failed at it (for lack of experience and resources) does not mean it should just be shrugged off as usel…

Only if misused. Our system supports human fallback, logging, and prompt tools to prevent poor user experiences. The key is thoughtful automation. GitHub: https://github.com/videosdk-live/agents Docs on HITL: https://docs.videosdk.live/ai_agents/human-in-the-loop

Re: Build an AI telephony agent for inbound and outbound calls

#43
post #15

Earlier quoted context omitted.

as somebody who works in this area: depends on how much effort they will be willing to invest. it might be better than current service or it might be worse. big question will be availability of humans as fall back that actually have some skills

I still can't wrap my head around how making functional online portals and well written documentation for customers is too hard, but laying off entire call centers and replacing them with a chat bot that relies on checks notes a functional set of APIs for tool use and well written documentation for a KB, is easier. It looks like all the same effort plus more!

That’s valid. But many people, including elderly users, prefer voice interfaces. Our system can serve those customers without requiring a smartphone or web access.

GitHub: https://github.com/videosdk-live/agents

Re: Build an AI telephony agent for inbound and outbound calls

#44
post #8

I'll take an alternative view of someone else's comment. No offense, but how much shittier will my customer service experience with my power company or ISP be because of this?

I'm someone who builds these types of voice agent systems for businesses. And the short answer is, I think especially for the bigger corporations, customer service probably will get shittier. It doesn't have to be inevitable...but the track record of big corps and their view of customer-facing departments as being loss-centers instead of being directly and indirectly connected to retaining and bringing in revenue doe…

That risk is real. That’s why we made this open-source to empower smaller businesses to build responsible systems with their own logic, prompts, and escalation paths. GitHub: https://github.com/videosdk-live/agents

Re: Build an AI telephony agent for inbound and outbound calls

#46
post #9

No offense, but how many people will be jobless because of this?

The question is totally fair, but it's unreasonable (imho) to expect owners of this project to have to answer. We are looking at My reaction was slightly different: how many companies selling this (meager) service at a high premium will go out of business now that it's free/open?

That is one of our goals. If a solution is under 500 lines of Python, you should not need to pay $499 per month for it. We want to lower the barrier for developers and businesses to build their own voice agents.

GitHub: https://github.com/videosdk-live/agents

https://docs.videosdk.live/ai_agents/voice-agent-quick-start

https://docs.videosdk.live/ai_agents/sip

Re: Build an AI telephony agent for inbound and outbound calls

#47

Just use pipecat

Thanks for the mention. Curious—what challenges are you finding with Pipecat that you're hoping something else (like https://github.com/videosdk-live/agents) might fix?

Always looking to improve based on real gaps devs are facing.

Re: Build an AI telephony agent for inbound and outbound calls

#48

Earlier quoted context omitted.

I would like to use AI to talk to those spammers/scammers. When the adversary is using AI why shouldn't we to defend from it.

You absolutely can. Our framework can answer calls, run speech-to-text, analyze intent, and respond with LLMs, making it a great defensive tool against spam or scam calls. GitHub: https://github.com/videosdk-live/agents

Great, I'll definitely check it out.

Re: Build an AI telephony agent for inbound and outbound calls

#49
post #26

Earlier quoted context omitted.

I did some freelancing around LLMs for startups and I can't tell you the amount of times I've had to reject "great ideas" that were just about spamming people. Almost had a big red sign "no outbound!"

Completely agree. Our goal is to enable smart automation for legitimate, helpful use cases, not intrusive marketing. We also built in Human-in-the-Loop (HITL) support so humans can intervene any time. Docs on HITL: https://docs.videosdk.live/ai_agents/human-in-the-loop GitHub: https://github.com/videosdk-live/agents

It's still illegal to place outbound automated calls, and it has been since like the 90s. There's even a simpsons episode about it
Post reply on HN