I hope they put beam forming into it. If it can't separate out real background noise, like TV, stereo, random chatter, as Google home, it's not going to be able to replace the minis. Thing is, I think they'd have to have 4 mics on that chip for it to perform that function. Hopefully PE is just to get it out there and the vNext will be more capable. I still have 2 on the way, maybe it will surprise me.
@@fedupgamer9075 so basically one the documentation page of wayogimg protocol there is a colab notebook You can clone it and and type your wake word in the script, provide training parameters and train for 3-5 hrs Once done it will create a file that file you have to add in shared or www folder don't remember exactly using samba Once done restart and your custom wake word will be in voice assistant menu from where you usually select between alexa , Jarvis and ok nabu
If this is using Bluetooth to connect to HA can you have multiple Voices so you can have one in each room? Totally love this premade solution. Far more elegant than I could ever make 😂
Not so far, as I know. I've heard that there is a small LLM included. Maybe thats why an Intel N100is recommended. But I'm sure you will add your ChatGPT Key sometimes. But this can never be run just local.
That is a great question. I would imagine if you can point to the media player entity, then I would think so. I don't have one of these yet so I can't test it, but being the speaker is a spearate media entity, I would think that would work. You will have to set up a Voice To Text engine and then write some custom script, but I would think it is doable. I look forward to hearing the true answer on this.
@@ggianne I did some quick searching and looked through the code and the media player is a separate entity for sure that I think can be called upon. Scripting will probably be required in order to have it say what you want when certain triggers happen as Home Assistant doesn't natuarlly know to "voice" everything that happens. I think it just needs to have text input in the script and then point it to a TTS engine and then output through the media player. However, I have not messed with any of that so there my very well be a better and easier way to do it.
Curious if you can add more than one of these devices to your smart home? I have 10 HomePods (Gen 2) and HomePod Minis scattered throughout my house. Can I do the same with the HAV PE?
@ honestly you and Chris Maher are the short form, easy to understand wled bible on TH-cam and a lot of knowledge would be wiped out if you guys went down!
Ah you're back lol. I thought someone had hacked TH-cam and was promoting crypto on everyone's feed until I asked some friends if they had such a thing on their home pages.
Using extended openai conversation add a openAI tokan or if you have power to run the models download from ollama or local AI and host on your network You will get your personalised assistance with a personality which uniquely yours Mine is clever but sometimes little passive aggressive and throws jokes here and there and sometimes brutally rostes people
I don't think Amazon is shaking in their boots just yet. You have dirty cheap Echos that provide decent speakers and can answer any smart home command or general question with little to no configuration. Don't get me wrong: this Voice Assistant is a nice step forward with lots of potential but remains for the geeks and tinkerers 😉
$60 for something you can get for $13? Also claims goals that are just goals of HA itself? nice. NO! Just get the ATOM Echo Smart Speaker Development Kit
Dude. Did you even think about your words? It is not only the hardware you are getting. You also support the ha creators. Ha is FREE. Do you expect that they work on it for free?
Welcome back!!
Thanks!! Glad things are getting set straight!
I hope they put beam forming into it. If it can't separate out real background noise, like TV, stereo, random chatter, as Google home, it's not going to be able to replace the minis. Thing is, I think they'd have to have 4 mics on that chip for it to perform that function.
Hopefully PE is just to get it out there and the vNext will be more capable.
I still have 2 on the way, maybe it will surprise me.
I never heard of that but I'll have to look into it!
Lee is know how it does for you
I‘d love to see you do the setup of the ESP32 S3 Box 3 und compare the two solutions.
Great! Hope it arrives soon!
I'd like to add some exotic Vocabulary, eg. Klingonic
....Imagine THAT wakup call 😅
I hope they allow you to use a custom wake word in the future.
@@fedupgamer9075 so basically one the documentation page of wayogimg protocol there is a colab notebook
You can clone it and and type your wake word in the script, provide training parameters and train for 3-5 hrs
Once done it will create a file that file you have to add in shared or www folder don't remember exactly using samba
Once done restart and your custom wake word will be in voice assistant menu from where you usually select between alexa , Jarvis and ok nabu
If this is using Bluetooth to connect to HA can you have multiple Voices so you can have one in each room? Totally love this premade solution. Far more elegant than I could ever make 😂
The bluetooth part is just for the first setup. After that. It will use wifi. So yes. You can have as many as you want.
Can you use this in combination with the ChatGPT API ?
Not so far, as I know. I've heard that there is a small LLM included. Maybe thats why an Intel N100is recommended.
But I'm sure you will add your ChatGPT Key sometimes. But this can never be run just local.
Hi! Can this device do some alerts like.. if thermometer temperature is higher than X then Home Assistant Voice say: (custom message set by user) ?
That is a great question. I would imagine if you can point to the media player entity, then I would think so. I don't have one of these yet so I can't test it, but being the speaker is a spearate media entity, I would think that would work. You will have to set up a Voice To Text engine and then write some custom script, but I would think it is doable. I look forward to hearing the true answer on this.
@a-a-ron21 let's hope it's easier... with Alexa i can setup this very easily
@@ggianne I did some quick searching and looked through the code and the media player is a separate entity for sure that I think can be called upon. Scripting will probably be required in order to have it say what you want when certain triggers happen as Home Assistant doesn't natuarlly know to "voice" everything that happens. I think it just needs to have text input in the script and then point it to a TTS engine and then output through the media player. However, I have not messed with any of that so there my very well be a better and easier way to do it.
This is really so cool.
Yeah! Gonna test it against the Esp32-box-3 and that new sat1
Curious if you can add more than one of these devices to your smart home? I have 10 HomePods (Gen 2) and HomePod Minis scattered throughout my house. Can I do the same with the HAV PE?
@@donpowers6033 Yes, you can. It was clearly mentioned during the presentation this week.
Yup!
Has it a speaker?
It sure does
Welcome back again lol
Thanks haha. I THINK everything is cleared up now haha
@ honestly you and Chris Maher are the short form, easy to understand wled bible on TH-cam and a lot of knowledge would be wiped out if you guys went down!
Thanks! Yeah I love his channel!
Ah you're back lol. I thought someone had hacked TH-cam and was promoting crypto on everyone's feed until I asked some friends if they had such a thing on their home pages.
Haha yeah that was bad! I think everything is finally back to normal!
not even close to Alexa
True! Alexa feeds all your data to Papa Bezos, HA Voice doesn't!
Using extended openai conversation add a openAI tokan or if you have power to run the models download from ollama or local AI and host on your network
You will get your personalised assistance with a personality which uniquely yours
Mine is clever but sometimes little passive aggressive and throws jokes here and there and sometimes brutally rostes people
I don't think Amazon is shaking in their boots just yet.
You have dirty cheap Echos that provide decent speakers and can answer any smart home command or general question with little to no configuration.
Don't get me wrong: this Voice Assistant is a nice step forward with lots of potential but remains for the geeks and tinkerers 😉
@@MakeItWorkTech that's true. But how much do I need to care if Amazon knows when I turn my lights on an off?
@michaelrch true, but I don't trust a microphone in my house I don't control. Still really like Alexa for dropping in, etc
$60 for something you can get for $13? Also claims goals that are just goals of HA itself? nice. NO! Just get the ATOM Echo Smart Speaker Development Kit
I'll definitely be testing that and comparing the two!
Dude. Did you even think about your words? It is not only the hardware you are getting. You also support the ha creators. Ha is FREE. Do you expect that they work on it for free?