I have a specific question: is it possible to create a chat flow within flowise AI where I could create one auto GPT supervisory agent and six auto GPT agents, all linked together in a process where each agent would have specific category of responsibility and the process would go through continuing iterations until the final outcome was acceptable? I believe this could be done with chains, but could it be done with autoGPT’s, which, for my application would be desirable in order to capture their greater reasoning capabilities?
Very good question probably you can get by by using agents and using different components of Lang chain to make something like that happen I am not sure how you would make it happen however I am pretty sure that there are easier tools to prop type such a complex solution, for example, autogen from Microsoft or crew AI much better processes to automate or to build a automation like the one that you are saying however great use case is great example and initially I would still product type it in lung chain with flow Wise so I have a MVP
Wow thanks a lot. Let me know if you have any ideas for a video. I would love to create a automation video or to automate the process that you would like to see. Thank you again and have a great day. Also if you would like to share any information with me like who you are where you’re from and if you wanna engage in a conversation, feel free to ping me on Linkedin.
You could use whisper from open AI or you can also use 11 lab the easiest would be to use open AI models because with one API you can do a lot of things
@@sullenfps I think you're a little behind the times. There are natural voice models out there that are indistinguishable from humans then by using an LLM GPU like Groq the response / latency down to milliseconds people will have no idea they are even speaking to AI.
There are no voice models on Flowise yet, but you can easily connect API endpoints to and from Deepgram, we use Bubble to create apps around flowise / LLM and have our endpoints connected to deepgram, actually are currently using VAPI for voice functionality but is has Deepgram support also.
@@RPAChampion Yes and soon when OpenAI finally open up voice function on GPT 4o via the api it will be quite a gamechanger as you wont have to make call to get voice input, then have the voice input transcribed into text via TTS model then through LLM to STT for output, which causes extra latency / response time. The whole prcoess will be contained within GPT 4o.........
Hey buddy, how are you doing? Are you checking out any AI automations? How’s work with my power? You are one of the first people to watch my channel thank you !
@@RPAChampion Moved away from Power Automate and have been in the GenAI/RPA/Agent space for automating IT Tier 1 Support calls. Looking forward to see GenAI videos in the future.
Good point I didn’t wanna make it extremely long, but I’m gonna make a more detail one about the system great and simple to make agents that can automate lots of different things quickly
![[LIVE] : ONE ลุมพินี 86 | คู่เอก "คมเพชร vs ชาติพยัคฆ์"](http://i.ytimg.com/vi/m3q_dRYDuU8/mqdefault.jpg)
Great video, thank you for being clear and consice!
Thank you very much for your comment. Have a great day and let me know if you have any ideas for future videos.
I have a specific question: is it possible to create a chat flow within flowise AI where I could create one auto GPT supervisory agent and six auto GPT agents, all linked together in a process where each agent would have specific category of responsibility and the process would go through continuing iterations until the final outcome was acceptable?
I believe this could be done with chains, but could it be done with autoGPT’s, which, for my application would be desirable in order to capture their greater reasoning capabilities?
Very good question probably you can get by by using agents and using different components of Lang chain to make something like that happen I am not sure how you would make it happen however I am pretty sure that there are easier tools to prop type such a complex solution, for example, autogen from Microsoft or crew AI much better processes to automate or to build a automation like the one that you are saying however great use case is great example and initially I would still product type it in lung chain with flow Wise so I have a MVP
Thanks!
Wow thanks a lot. Let me know if you have any ideas for a video. I would love to create a automation video or to automate the process that you would like to see. Thank you again and have a great day. Also if you would like to share any information with me like who you are where you’re from and if you wanna engage in a conversation, feel free to ping me on Linkedin.
great video! Can I bring other AI models such as from deepgram ? I am trying to build coversational ai phone agent. Thank you
nobody wants to talk to ai on the phone
You could use whisper from open AI or you can also use 11 lab the easiest would be to use open AI models because with one API you can do a lot of things
@@sullenfps I think you're a little behind the times. There are natural voice models out there that are indistinguishable from humans then by using an LLM GPU like Groq the response / latency down to milliseconds people will have no idea they are even speaking to AI.
There are no voice models on Flowise yet, but you can easily connect API endpoints to and from Deepgram, we use Bubble to create apps around flowise / LLM and have our endpoints connected to deepgram, actually are currently using VAPI for voice functionality but is has Deepgram support also.
@@RPAChampion Yes and soon when OpenAI finally open up voice function on GPT 4o via the api it will be quite a gamechanger as you wont have to make call to get voice input, then have the voice input transcribed into text via TTS model then through LLM to STT for output, which causes extra latency / response time. The whole prcoess will be contained within GPT 4o.........
Gotta rebrand to AI Champion.
Hey buddy, how are you doing? Are you checking out any AI automations? How’s work with my power? You are one of the first people to watch my channel thank you !
@@RPAChampion Moved away from Power Automate and have been in the GenAI/RPA/Agent space for automating IT Tier 1 Support calls.
Looking forward to see GenAI videos in the future.
Thanks for the video. But you missed the whole automation part
Good point I didn’t wanna make it extremely long, but I’m gonna make a more detail one about the system great and simple to make agents that can automate lots of different things quickly