Thanks for the Wyoming protocol overview. That demystifies things a little bit for me. I would like to see more on this topic in the future if you have time. I run piper and whisper in a docker container along with ha and other add-ons. Going over this type setup would be helpful to others who run a similar environment. Thank you for taking the time to share. Joe
You're very welcome 😊. And as i've written on the other comment, i'm playing around with docker instances and wyoming and plan to make a more detailed video about it. This offers really cool possibilities.
I would like to receive instructions from your channel for the reverse task, how to use this protocol to work with other TTS projects. in order to integrate third-party tts into the Assist's pipeline
@@ThorstenMueller Thank you for your time. I would like to know how to create a wyoming server that interacts with the tts model (as done in the wyoming-glados project). Personally, I am interested in integrating the local vosk-tts project (currently only ru). But the explanations can be demonstrated on any similar tts with api.
I have a question for you. I saw a video where an ai used some type of text to speech to sing a song with a monslisa face. Can piper tts do that. I tried using piper tts to read the king james bible and I found some issues with pronouncing some words. The king james bible is the ultimate test for any text to speech software. If it is possible to make piper tts sing them can you me how to do it. Thanks in advance.
Piper TTS models can not sing AFAIK and doing face sync is out of scope for a pure TTS solution. But maybe it works with Piper in combination with other AI tools.
Lieber Thorsten, vielen Dank für deine Videos! Ich habe einige davon angesehen, aber als Pfälzer finde ich sie schwer zu verstehen, :) Mein Pfälzisch ist sehr ausgeprägt, und ich denke, es würde sich großartig anhören, wenn ich Videos in eine andere menschliche Sprache vertonen könnte. Habe meine Stimme erfolgreich mithilfe von Chat-GPT übersetzt, und es funktioniert wirklich sehr gut. Allerdings bin ich mir nicht sicher, ob ich diese Stimmen für TH-cam verwenden darf. Ich habe bereits an Chat-GPT geschrieben, aber bisher keine Antwort erhalten. Deshalb suche ich derzeit nach Alternativen, die ich legal verwenden kann. Kannst du mir dabei helfen?
Da ich keine juristischen Kenntnisse habe, kann ich dazu nichts genaues sagen. Ich habe auch mal in deine Videos geschaut und verstehe, was du mit dem ausgeprägten Pfälzer Dialekt meinst 😉.
@@jason2679 AFAIK Piper TTS does not work on mobile devices (locally), but maybe you can run a publically available webservice for TTS and use it from your mobile device.
I am desperate to get rid of Alexa voice control of my HA server. I can't understand why Nabu Casa simply doesn't develop a speaker/phone hardware device that works out of the box with the new HA local voice. It would sell like hotcakes. Do we even know if they are developing one..say with this Wyoming protocol??
you sir, are awesome. thank you. I couldn't make heads or tails of the sequence of installing things. Also love the fact that you can create a new VM/container specifically for this! #easySubscribe
I want to make an AI voicebot and I am using windows, so I want to use piper in python on windows. Can you please help me out or can you please one complete video on how to use piper TTS in python on windows
Hi @Thorsten-Voice, thank a lot for all videos, they helped me a lot! Question for you, I want to train model from scratch (I know it will take a lot of time etc), could you tell me what is the best open source TTS project to use for this, since there are a lot of them? Piper, coqui-ai etc.Also could you tell me in your experience, how big the data set should be for a proof of concept (meaning it can export simple sentence) and for the best results possible (any text provided should have decent result in export)? I would like just to get some kind of proof of concent first, since I do not have nvidia, i only have laptop with 8/16 cores/threads, and amd gpu. Thanks a lot and thanks for your work!
I'd go with Piper TTS as it has active development and is really fast and lightwight (no huge dependencies). Maybe start with 500-1000 recordings for a proof of concept.
Thanks for the Wyoming protocol overview. That demystifies things a little bit for me. I would like to see more on this topic in the future if you have time. I run piper and whisper in a docker container along with ha and other add-ons. Going over this type setup would be helpful to others who run a similar environment. Thank you for taking the time to share.
Joe
You're very welcome 😊. And as i've written on the other comment, i'm playing around with docker instances and wyoming and plan to make a more detailed video about it. This offers really cool possibilities.
I would like to receive instructions from your channel for the reverse task, how to use this protocol to work with other TTS projects. in order to integrate third-party tts into the Assist's pipeline
You mean adding wyoming support/compatibility to other tts solutions? Do you have a special tts solution in mind?
@@ThorstenMueller Thank you for your time. I would like to know how to create a wyoming server that interacts with the tts model (as done in the wyoming-glados project). Personally, I am interested in integrating the local vosk-tts project (currently only ru). But the explanations can be demonstrated on any similar tts with api.
I have a question for you. I saw a video where an ai used some type of text to speech to sing a song with a monslisa face. Can piper tts do that. I tried using piper tts to read the king james bible and I found some issues with pronouncing some words. The king james bible is the ultimate test for any text to speech software. If it is possible to make piper tts sing them can you me how to do it. Thanks in advance.
Piper TTS models can not sing AFAIK and doing face sync is out of scope for a pure TTS solution. But maybe it works with Piper in combination with other AI tools.
@@ThorstenMueller I was just wondering since I saw a youtube video on this. It would have been wonderful if piper tts could have done that.
Like Share Die
Also dies Mal hatte ich timing , erster by the way
Nice content. Thx and like
i'd love more developer tutorial videos for the wyoming protocol
Thanks for your suggestion, i've added it to my growing TODO list 😊.
Lieber Thorsten,
vielen Dank für deine Videos! Ich habe einige davon angesehen, aber als Pfälzer finde ich sie schwer zu verstehen, :) Mein Pfälzisch ist sehr ausgeprägt, und ich denke, es würde sich großartig anhören, wenn ich Videos in eine andere menschliche Sprache vertonen könnte. Habe meine Stimme erfolgreich mithilfe von Chat-GPT übersetzt, und es funktioniert wirklich sehr gut. Allerdings bin ich mir nicht sicher, ob ich diese Stimmen für TH-cam verwenden darf. Ich habe bereits an Chat-GPT geschrieben, aber bisher keine Antwort erhalten. Deshalb suche ich derzeit nach Alternativen, die ich legal verwenden kann. Kannst du mir dabei helfen?
Da ich keine juristischen Kenntnisse habe, kann ich dazu nichts genaues sagen. Ich habe auch mal in deine Videos geschaut und verstehe, was du mit dem ausgeprägten Pfälzer Dialekt meinst 😉.
Can make video how I can use TTS on more ? For ai characters
What do you mean with "TTS on more"?
@@ThorstenMueller I want to use TTS in my mobile so please tell me how can I use it
@@jason2679 AFAIK Piper TTS does not work on mobile devices (locally), but maybe you can run a publically available webservice for TTS and use it from your mobile device.
I am desperate to get rid of Alexa voice control of my HA server. I can't understand why Nabu Casa simply doesn't develop a speaker/phone hardware device that works out of the box with the new HA local voice. It would sell like hotcakes. Do we even know if they are developing one..say with this Wyoming protocol??
I have no inside to the plans of Nabu Casa, but they provide or work together with these esp room devices. I am excited about where this is going.
you sir, are awesome. thank you. I couldn't make heads or tails of the sequence of installing things.
Also love the fact that you can create a new VM/container specifically for this!
#easySubscribe
Thanks a lot for your very kind feedback and welcome to my community 🤩.
I want to make an AI voicebot and I am using windows, so I want to use piper in python on windows. Can you please help me out or can you please one complete video on how to use piper TTS in python on windows
I added this special topic "Piper TTS in python on windows" to my TODO list 😊.
Just found your channel and I'm soaking up all I can. Thanks for your efforts in this category, it's much appreciated.
Thanks a lot for that great feedback 😊.
Hi @Thorsten-Voice, thank a lot for all videos, they helped me a lot! Question for you, I want to train model from scratch (I know it will take a lot of time etc), could you tell me what is the best open source TTS project to use for this, since there are a lot of them? Piper, coqui-ai etc.Also could you tell me in your experience, how big the data set should be for a proof of concept (meaning it can export simple sentence) and for the best results possible (any text provided should have decent result in export)? I would like just to get some kind of proof of concent first, since I do not have nvidia, i only have laptop with 8/16 cores/threads, and amd gpu. Thanks a lot and thanks for your work!
I'd go with Piper TTS as it has active development and is really fast and lightwight (no huge dependencies). Maybe start with 500-1000 recordings for a proof of concept.
@@ThorstenMueller thank you a lot