• Reminder: Do not call, text, or mention harrassing someone in real life. Do not encourage it. Do not talk about killing or using violence against anyone, or engaging in any criminal behavior. If it is not an obvious joke even when taken out of context, don't post it. Please report violators.

    DMCA, complaints, and other inquiries:

    [email protected]

O&A AI reunion show

Dr. Adam Cox

Dr. Adam Cox

Yes, I'm trying to raise super kids
‎
Forum Clout
927
Uncle J’s Sink Emporium said:
Incredible. Someone smart could probably write some kind of program that would scrape the news each morning and make a fake show out of it.
Click to expand...
I thought about that a while ago. It's possible, but it's a lot of work, especially if you want their dialogue to actually sound somewhat natural and accurate. Here's the high level flow I came up with:
  1. The program starts by calling GPT4 for a list of current events, with a focus on topics relevant to what O&A would discuss, and output it to some format.
  2. Now you could use GPT-4 to create the actual script, but it won't have enough data to output something that would sound like them actually conversing. To get that, you could use a local LLM and train a LoRA on a dataset full of quotes and conversations from O&A and certain facts that might not be there already, like nana's love for a boy named sue.
  3. Send the list of events and stuff to the local model with a prompt for creating a script that will output to a standard format. Depending on how long you want it to be, the token limit of the model, and the hardware used to run the model, you may have to break it up into separate prompts.
  4. Then you have the app break up the conversation and send each instance of each speaker talking to a text-to-speech program (this can also be done locally).
  5. Most TTS models aren't very accurate, so you'd probably then want to export the audio files to RVC (which was used for OP's audio) to replace the TTS voices with O&A's. The files should output with some sequenced naming convention.
  6. The program could then re-compile the conversation into one single audio file and throw Street Fighting Man at the beginning.
The only problem here would be spacing out the audio files when creating the combined audio track. I have no idea how you could automate natural pauses and interruptions, so you'd probably be better off manually editing it all together, but that's still not super easy to do.
There's a lot of manual stuff that would make it better, like recording yourself reading the script and using RVC on that instead of the TTS output, but this is just a theoretical automated end to end solution.

so good luck to to whoever wants to make that
 
Last edited:
Riccardo Bosi

Riccardo Bosi

has janny powers
‎
‎
Forum Clout
70,612
Dr. Adam Cox said:
I thought about that a while ago. It's possible, but it's a lot of work, especially if you want their dialogue to actually sound somewhat natural and accurate. Here's the high level flow I came up with:
  1. The program starts by calling GPT4 for a list of current events, with a focus on topics relevant to what O&A would discuss, and output it to some format.
  2. Now you could use GPT-4 to create the actual script, but it won't have enough data to output something that would sound like them actually conversing. To get that, you could use a local LLM and train a LoRA on a dataset full of quotes and conversations from O&A and certain facts that might not be there already, like nana's love for a boy named sue.
  3. Send the list of events and stuff to the local model with a prompt for creating a script that will output to a standard format. Depending on how long you want it to be, the token limit of the model, and the hardware used to run the model, you may have to break it up into separate prompts.
  4. Then you have the app break up the conversation and send each instance of each speaker talking to a text-to-speech program (this can also be done locally).
  5. Most TTS models aren't very accurate, so you'd probably then want to export the audio files to RVC (which was used for OP's audio) to replace the TTS voices with O&A's. The files should output with some sequenced naming convention.
  6. The program could then re-compile the conversation into one single audio file and throw Street Fighting Man at the beginning.
The only problem here would be spacing out the audio files when creating the combined audio track. I have no idea how you could automate natural pauses and interruptions, so you'd probably be better off manually editing it all together, but that's still not super easy to do.
There's a lot of manual stuff that would make it better, like recording yourself reading the script and using RVC on that instead of the TTS output, but this is just a theoretical automated end to end solution.

so good luck to to whoever wants to make that
Click to expand...
tenor.gif
 
1073waaf

1073waaf

The ONLY station that REALLY ROCKS!
‎
‎
Forum Clout
20,467
Dr. Adam Cox said:
I thought about that a while ago. It's possible, but it's a lot of work, especially if you want their dialogue to actually sound somewhat natural and accurate. Here's the high level flow I came up with:
  1. The program starts by calling GPT4 for a list of current events, with a focus on topics relevant to what O&A would discuss, and output it to some format.
  2. Now you could use GPT-4 to create the actual script, but it won't have enough data to output something that would sound like them actually conversing. To get that, you could use a local LLM and train a LoRA on a dataset full of quotes and conversations from O&A and certain facts that might not be there already, like nana's love for a boy named sue.
  3. Send the list of events and stuff to the local model with a prompt for creating a script that will output to a standard format. Depending on how long you want it to be, the token limit of the model, and the hardware used to run the model, you may have to break it up into separate prompts.
  4. Then you have the app break up the conversation and send each instance of each speaker talking to a text-to-speech program (this can also be done locally).
  5. Most TTS models aren't very accurate, so you'd probably then want to export the audio files to RVC (which was used for OP's audio) to replace the TTS voices with O&A's. The files should output with some sequenced naming convention.
  6. The program could then re-compile the conversation into one single audio file and throw Street Fighting Man at the beginning.
The only problem here would be spacing out the audio files when creating the combined audio track. I have no idea how you could automate natural pauses and interruptions, so you'd probably be better off manually editing it all together, but that's still not super easy to do.
There's a lot of manual stuff that would make it better, like recording yourself reading the script and using RVC on that instead of the TTS output, but this is just a theoretical automated end to end solution.

so good luck to to whoever wants to make that
Click to expand...
With the weekly windows 10 updates, I've had to have my OS reinstalled TWICE due to compatibility issues. Now, $200 geek squad dollars later, I've eliminated that mess but I get this pop-up scam, when I take the computer out of sleep mode, that has a window that won't close and audio of some limey twunt telling me that my computer's been infected by a virus from a malicious porn site and that I need to call the number on the screen because all of my personal info including Passwords for banking, social media sites, credit cards, etc... etc... have been compromised.

Most asshats would shit themselves hearing this blaring from their computer speakers, as the very official looking screen blinks an 1-888 number and implores you to call it to work with their "windows professionals" to eliminate this infection.

If this happens to you, do the following:

Hold Control, alt, delete simultaneously.

Choose "task manager" from the resulting menu, and close your browser.

It disappears. I'm not sure if it'll happen again, but for Christ's sake... DON'T call the number and play ball with these mongrel terrorists. The only infection your computer has is some fucker slipped in that pop up in the form of a familiar email with a link you (and I) mistakenly clicked.

Go to "remove program" in your control panel and look for one of these. Delete it, problem solved. THANKS FOR YOUR USELESS SECURITY UPDATES WINDOWS 10!!!!!!
 
BonnieMcFarlaneMe2

BonnieMcFarlaneMe2

❤️subscribe to the bon-fire❤️
‎
‎
Forum Clout
90,528
Dr. Adam Cox said:
I thought about that a while ago. It's possible, but it's a lot of work, especially if you want their dialogue to actually sound somewhat natural and accurate. Here's the high level flow I came up with:
  1. The program starts by calling GPT4 for a list of current events, with a focus on topics relevant to what O&A would discuss, and output it to some format.
  2. Now you could use GPT-4 to create the actual script, but it won't have enough data to output something that would sound like them actually conversing. To get that, you could use a local LLM and train a LoRA on a dataset full of quotes and conversations from O&A and certain facts that might not be there already, like nana's love for a boy named sue.
  3. Send the list of events and stuff to the local model with a prompt for creating a script that will output to a standard format. Depending on how long you want it to be, the token limit of the model, and the hardware used to run the model, you may have to break it up into separate prompts.
  4. Then you have the app break up the conversation and send each instance of each speaker talking to a text-to-speech program (this can also be done locally).
  5. Most TTS models aren't very accurate, so you'd probably then want to export the audio files to RVC (which was used for OP's audio) to replace the TTS voices with O&A's. The files should output with some sequenced naming convention.
  6. The program could then re-compile the conversation into one single audio file and throw Street Fighting Man at the beginning.
The only problem here would be spacing out the audio files when creating the combined audio track. I have no idea how you could automate natural pauses and interruptions, so you'd probably be better off manually editing it all together, but that's still not super easy to do.
There's a lot of manual stuff that would make it better, like recording yourself reading the script and using RVC on that instead of the TTS output, but this is just a theoretical automated end to end solution.

so good luck to to whoever wants to make that
Click to expand...
You lost me somewhere between 1 & 2.
 
Rick Roblinson

Rick Roblinson

Sausage-Fingered Vulgarian
‎
Forum Clout
19,925
Dr. Adam Cox said:
I thought about that a while ago. It's possible, but it's a lot of work, especially if you want their dialogue to actually sound somewhat natural and accurate. Here's the high level flow I came up with:
  1. The program starts by calling GPT4 for a list of current events, with a focus on topics relevant to what O&A would discuss, and output it to some format.
  2. Now you could use GPT-4 to create the actual script, but it won't have enough data to output something that would sound like them actually conversing. To get that, you could use a local LLM and train a LoRA on a dataset full of quotes and conversations from O&A and certain facts that might not be there already, like nana's love for a boy named sue.
  3. Send the list of events and stuff to the local model with a prompt for creating a script that will output to a standard format. Depending on how long you want it to be, the token limit of the model, and the hardware used to run the model, you may have to break it up into separate prompts.
  4. Then you have the app break up the conversation and send each instance of each speaker talking to a text-to-speech program (this can also be done locally).
  5. Most TTS models aren't very accurate, so you'd probably then want to export the audio files to RVC (which was used for OP's audio) to replace the TTS voices with O&A's. The files should output with some sequenced naming convention.
  6. The program could then re-compile the conversation into one single audio file and throw Street Fighting Man at the beginning.
The only problem here would be spacing out the audio files when creating the combined audio track. I have no idea how you could automate natural pauses and interruptions, so you'd probably be better off manually editing it all together, but that's still not super easy to do.
There's a lot of manual stuff that would make it better, like recording yourself reading the script and using RVC on that instead of the TTS output, but this is just a theoretical automated end to end solution.

so good luck to to whoever wants to make that
Click to expand...
HOLY S
 
SicilanBrotherMan

SicilanBrotherMan

Not a fan of that Fatrick guy
‎
Forum Clout
262
Dr. Adam Cox said:
I thought about that a while ago. It's possible, but it's a lot of work, especially if you want their dialogue to actually sound somewhat natural and accurate. Here's the high level flow I came up with:
  1. The program starts by calling GPT4 for a list of current events, with a focus on topics relevant to what O&A would discuss, and output it to some format.
  2. Now you could use GPT-4 to create the actual script, but it won't have enough data to output something that would sound like them actually conversing. To get that, you could use a local LLM and train a LoRA on a dataset full of quotes and conversations from O&A and certain facts that might not be there already, like nana's love for a boy named sue.
  3. Send the list of events and stuff to the local model with a prompt for creating a script that will output to a standard format. Depending on how long you want it to be, the token limit of the model, and the hardware used to run the model, you may have to break it up into separate prompts.
  4. Then you have the app break up the conversation and send each instance of each speaker talking to a text-to-speech program (this can also be done locally).
  5. Most TTS models aren't very accurate, so you'd probably then want to export the audio files to RVC (which was used for OP's audio) to replace the TTS voices with O&A's. The files should output with some sequenced naming convention.
  6. The program could then re-compile the conversation into one single audio file and throw Street Fighting Man at the beginning.
The only problem here would be spacing out the audio files when creating the combined audio track. I have no idea how you could automate natural pauses and interruptions, so you'd probably be better off manually editing it all together, but that's still not super easy to do.
There's a lot of manual stuff that would make it better, like recording yourself reading the script and using RVC on that instead of the TTS output, but this is just a theoretical automated end to end solution.

so good luck to to whoever wants to make that
Click to expand...
16 internet dick
 
You must log in or register to reply here.
Top