Having an Actual Conversation with Data Using an ElevenLabs Voice Agent and n8n

Поделиться
HTML-код
  • Опубликовано: 2 фев 2025

Комментарии • 29

  • @carlocatabijan
    @carlocatabijan 21 день назад +2

    How it pauses to breathe, intonate points, and have natural spaces in talking is insane.

  • @benoittoutant5027
    @benoittoutant5027 20 дней назад

    The Elenven Labs voice is a real game changer.

  • @CatsInHatz
    @CatsInHatz 21 день назад +1

    Great video - it would be fantastic to have a tutorial with something like Aircall acting as a receptionist. Is this something easy enough?
    1. Call number -> trigger webhook
    2. Back and forth -> who do you need to speak with/Basic answer questions
    3. Says sales person name/department.
    4. Forward to correct person/department via a webhook

  • @eduardmedina4257
    @eduardmedina4257 21 день назад +1

    Awesome!!!

  • @pourkin
    @pourkin 17 дней назад

    excellent tutorial

  • @agustindesimonne
    @agustindesimonne 21 день назад

    So interesting Nate! I wonder how do you come up with these ideas, they're so creative

  • @connormartin2601
    @connormartin2601 21 день назад

    Thanks mate

  • @KartikSehgal-b7t
    @KartikSehgal-b7t 8 дней назад

    Can you make a video explaining how we can connect GHL voice agent with database in n8n, Thanks

  • @Jasmin-Ramsy0
    @Jasmin-Ramsy0 21 день назад +1

    Can you create reservation google calendar elevenlabs voice agent

  • @natewhite6922
    @natewhite6922 19 дней назад

    Curious, new to n8n, worked great on test once, and the whole workflow showed the data from my data base in the end webhook node, i went to the production URL (activated the flow) - he couldn't have access to the data base. Then one time it worked! but mostly doesn't work.

  • @PyJu80
    @PyJu80 19 дней назад

    Do you do this as a tease to make us join your channel? Because I followed everything, but my workflow on n8n doesnt trigger, also where do I get the data for the project ?Slow down too. It's soooo hard for begginers to follow along, some of us have never used any of these tools. I did test and production, followed everything three times, and all I got was to talk to the voice agent, nothing in n8n worked. Bro....

  • @bangkitsanjaya3773
    @bangkitsanjaya3773 18 дней назад

    Supabase and pinecone, which one better? My vector store from supabase table is always miss documents when retrieve 😢

    • @nateherk
      @nateherk  17 дней назад

      It's hard to say which one is better, it depends on the use case and the type of data. Also depends on factors like hosting and scalability.

  • @holylittlegirl
    @holylittlegirl 20 дней назад

    There is no agent memory, which is similar to my issue. How can I give memory to this agent?
    With VAPI, it's easy, but the same voice sounds different (11labs sounds much better), in my opinion.

  • @GregDolan
    @GregDolan 13 дней назад

    Great implimentation, wish you went slower and showed how to build it step by step though. This was more of a tease 🙃

  • @petersonsouza9600
    @petersonsouza9600 21 день назад

    Incredible!! I will try that one. Thanks for sharing

  • @shubh9207
    @shubh9207 21 день назад

    Nuts

  • @JamesMartin2014
    @JamesMartin2014 21 день назад +5

    how tf is yours so fast? My n8n is slow af. I thought it was the openai api.....

    • @tal7atal7a66
      @tal7atal7a66 21 день назад

      Maybe you have to upgrade your instance ( i think n8n need 2 cpu , 4gb ram, for basic use cases )

  • @tal7atal7a66
    @tal7atal7a66 21 день назад

    Elevenlabs or vapi wich one better ? 🤔

    • @nateherk
      @nateherk  20 дней назад

      That’s the real question!

  • @BerkcanHafizoglu
    @BerkcanHafizoglu 20 дней назад +1

    I couldn't replicate this workflow for some reason. I followed all the steps you did, but the agent still doesn't work. It would be great if you could dive deeper into the technical aspects to help us understand better. Since I can't fully grasp what's happening, I'm unable to fix it.

    • @nateherk
      @nateherk  20 дней назад +2

      Hey, thanks for the feedback. I can look at doing a deeper dive to make sure you can get unstuck with this workflow!

    • @BerkcanHafizoglu
      @BerkcanHafizoglu 20 дней назад

      @@nateherk thanks Nate!

  • @RomuloMagalhaesAutoTOPO
    @RomuloMagalhaesAutoTOPO 20 дней назад

  • @greendsnow
    @greendsnow 20 дней назад +1

    yeah but how do you validate the user? you're not opening your secrets to the whole internet, are you?

    • @troymcdonald
      @troymcdonald 18 дней назад

      Well, you’d not do this kind of setup for highly sensitive information. For highly sensitive information I’d say you’d have to have a password. You’d be able to have a password for each person.