I agree , been searching for sometime for something like that the flexblity of real time chat with local tts&stt, flowise or 8n8 or even open webui without using thier template ui
Is this possible with Flowise? Would be great to see a tutorial of AI Voice Agents with Flowise! There are many options there: assembly, elevenlabs, deepgram.. but not sure which one to use within Flowise.
Excellent like always! The content you create is probably the most valuable for how to apply ai in real cases! It’s awesome to focus on creating in profuction ready apps that use AI. I know that you may be to busy (and it can be very complicated to develop) but just in case you look for more ideas: creating an agent that can handle files processing and geneating, with also web scraping and data entry. Also with the possibility of doing it in a server (so it can run in the background). I know this could be an extremely (even impossible) task. But just in case you know how to do it, it would be the best ai use case ever i think
Thanks Leon for sharing this. Got stuck with multi agent langchain and openai, getting the rights packages makes me nuts. Will move to crewai and will add this :). Really awesome.
So, funny enough they do provide callback functions that return the messages (you can actually see it in the final code in my github repo). I didn't mention it in the video because at the time of recording it didn't seem to actually work all that well. To be fair, the SDK is still in beta, but I wanted to include it in the final code for reference either way.
I have the exact same code adjusted for React, but I'm not hearing anything after starting the conversation. "Connected to ElevenLabs" gets logged, and then "Started conversation:" with convo ID, but even when I speak there's no response and it disconnects on it own after a few seconds of silence. What could be the issue?
![[Urdu/Hindi] How to set up a new Next js project | Tutorial 2](/img/n.gif)
What's the most creative way you'd use a real-time voice agent in your Next.js app?
would love to see a video on connecting this in a n8n or flowise workflow for a voice/text chatbot use case
I agree , been searching for sometime for something like that the flexblity of real time chat with local tts&stt, flowise or 8n8 or even open webui without using thier template ui
Is this possible with Flowise? Would be great to see a tutorial of AI Voice Agents with Flowise! There are many options there: assembly, elevenlabs, deepgram.. but not sure which one to use within Flowise.
That would be awesome!! Hoping FW integrates this into the chat embed soon.
Excellent like always! The content you create is probably the most valuable for how to apply ai in real cases!
It’s awesome to focus on creating in profuction ready apps that use AI.
I know that you may be to busy (and it can be very complicated to develop) but just in case you look for more ideas: creating an agent that can handle files processing and geneating, with also web scraping and data entry. Also with the possibility of doing it in a server (so it can run in the background).
I know this could be an extremely (even impossible) task. But just in case you know how to do it, it would be the best ai use case ever i think
Thank you very much for the feedback! I'm always looking for new video ideas, so appreciate it.
Thanks Leon for sharing this. Got stuck with multi agent langchain and openai, getting the rights packages makes me nuts. Will move to crewai and will add this :). Really awesome.
You're welcome
Try flowise, 1000% simpler
Very cool!! Thx!! Can you tell how to let an iPhone connected to the host to actually work? I cannot find a way to let the hardware work
great fan here
Just glad my videos add value
That’s awesome. Thanks for sharing
You're welcome! 😊
Hey, Just curious, can we add a audio visualizer for user and ai voice like round bars or something like that ?
Noice! Would be nice to include a transcript chat window beside it that includes stt for both parties in the template 😎
So, funny enough they do provide callback functions that return the messages (you can actually see it in the final code in my github repo).
I didn't mention it in the video because at the time of recording it didn't seem to actually work all that well. To be fair, the SDK is still in beta, but I wanted to include it in the final code for reference either way.
Amazing,
Is there anyway to get it done using open source tools ?? 🤔
Appreciate your videos
I have the exact same code adjusted for React, but I'm not hearing anything after starting the conversation. "Connected to ElevenLabs" gets logged, and then "Started conversation:" with convo ID, but even when I speak there's no response and it disconnects on it own after a few seconds of silence. What could be the issue?
Do you hear anything when testing via the elevenlabs dashboard? You might need to change your output device in your browser
@@leonvanzyl I used a different agent ID and it worked, thanks for replying though!
Great work, I once suggested that you could do that and thank you for reminding me 😅
Hahaha, you're welcome 🤗
Thhank you
You're welcome 🤗
Flippin cool, dankie Leon 🙏 enige moontlike nuwe FW tips and tricks vids voor XMaS? 🤭
Verseker! Nuwe FW kom more uit.
Woow you guys speak Afrikaans? As a dutch person it is so weird reading something weird and then realizing I can make sense of it 😂😊
@@NS_Miata i was also surprised when i visited Antwerpen how similar Afrikaans is to Flemish … 👍 “Leuk je te ontmoeten!”
Love this! Off to play 😁😁 any idea of baseline costs per minute..
It seems to be free during the beta phase, so I don't think they've bedded down the pricing. Check out the pricing page though
Can I use my own voice?
Yup!
Very cool!! Thx!! Can you tell how to let an iPhone connected to the host to actually work? I cannot find a way to let the hardware work