Activity
Mon
Wed
Fri
Sun
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
What is this?
Less
More
Memberships
Open Source Voice AI Community
763 members • Free
AI Automation Security
61 members • Free
Voice AI Accelerator
6.6k members • Free
Learn Voice AI
375 members • Free
IA al teléfono
78 members • $39/m
AI Automation Society
206.4k members • Free
7 contributions to Open Source Voice AI Community
21d •
Best Observability Tools for Voice AI Frameworks?
What observability tools are others using with Pipecat or similar voice AI frameworks? I've built a production voice agent using Pipecat and currently track basic metrics (call duration, sentiment, summary, transcripts) in a custom dashboard. Tomorrow it's going in production so problem I think I can face is When errors will occur, debugging is painful. My current logging approach creates massive log files that are nearly impossible to analyze efficiently when tracking down issues.
0 likes • 21d
@Mohammad Mussab I've just started using Logfire from Pydantic (https://pydantic.dev/logfire). I'm working on a whatsapp chatbot at the moment so not tested it with voice but it's so handy to quickly identify errors through their logs. Will be checking out Whisker though for sure.
23d •
Just sayin
Honestly, I've been building voice AI agents with Ultravox AI and the help of Claude. I understand that this community is really about LiveKit and Pipecat and open source, but Ultravox is also open source. I've had a lot of success using Ultravox AI. I've had some success with LiveKit and Pipecat, but I've had the most success with Ultravox AI. I think it's undervalued and overlooked as a source for open source AI agents and building them.oh, and…Claude is king!
1 like • 22d
@Robert Figueroa Loving Claude Code too - (except the token limits - $20 to $100 is a big jump!).
1 like • 22d
@Robert Figueroa Not using much in the way of sub-agents yet. But I did introduce Serena recently which apparently helps reduce token use significantly.
27d •
Whitelabel Dashboard for Agencies for Scale and Success
We just launched our whitelabel dashboard on Convolytic.com for agencies and service providers. Here you can share analysis and create a custom domain for your clients to see the success of their business agent. We also allow you to track and bill via minutes used. We will be also adding deeper business insights there to help them unlock revenue opportunities. If you're interested let me know, I'm looking for VoiceAI developers to try it for 6 WEEKS for free and give me feedback.
1 like • 26d
@Mohammad Mussab Not everyone wants to make their own dashboards. If they did, there wouldn't be companies like ChatDash. It's a different business model entirely. And it's a learning curve. For every minute you are building a dashboard, you aren't building or selling a voice agent. That being said, I'm making my own too:)
26d •
🤔 What’s the right number of tools for a single AI agent?
As I’m building a front-desk agent for clinics & restaurants, my current setup uses one agent with tools like: - booking - cancel - reschedule - get_current_time - save_context_in_redis I want to know from people who’ve built production systems: Is there an ideal number of tools a single agent should handle? Or is it better to split this into multiple smaller agents?
1 like • 26d
@Abdulbasit Arif Your agent should be able to comfortably manage those 5 tools. You can go higher but either way you'll need a rock solid prompt, excellent tool descriptions and some examples of use for each tool.
29d •
N8N for production?
I was wondering if n8n is suitable for production in terms of reliability and scalability, can we build a reliable AI voice platform based on n8n as backend? Or just for demos? Here I’m talking about multi-tenant platforms where we can onboard multiple clients. And What about n8n license, do we need a license for using n8n commercially inside our platform/applications?
1 like • 29d
@Ahmed Elmidani I think a lot depends on how you intend to use n8n in your back end and the scale. For example, you might want to use it as part of the onboarding process, or to move data from one source to another. It might also depend on your existing infrastructure and the expertise in your current team. Of course, there are pros and cons, but there's certainly a use case for n8n if we think about our Voice AI platforms being part of a wider Voice AI system.
1-7 of 7