mirror of
https://github.com/tinygrad/tinygrad.git
synced 2026-01-07 22:23:55 -05:00
5e87083783
* feat: working voice 2 text using whisper * feat: added llama generation * feat: vits init * feat: more accurate voice conversion * feat: support for tts and working pipeline for the first pass * fix: linter checks * refactored vits initialization and inference, added mmts-tts support * fixed process sync and now we can have an infinite conversation * reuse output stream to remove overhead of creating a new one each time * added pre-prompt configuration with yaml files * adjusted code to merge PR which changed whisper * optimized whisper, now it's blazing fast and also reduced number of lines * added better debug printing * use jitted encode function for whisper, added timings and removed response delim to save speed on generating those tokens * fixed hf convert and now it's working with tinyllama * added tinyllama config * refactored code and made it work with all llama models * prettier order * prettier order * fixed suffix for tinyllama and refactored convert_from_hf * added missing parameters * fixed stream release and added missing params * jitted dp and encoder * jitted flow forward * removed re-init of espeak on each call to save up time * jitted generator forward for blazing fast tts * added contextmanager for displaying a chat log * removed whitespace for pylint * updated code to support latest fetch func * wait for llama eos token and pass params from cli to llama * listen for not fixed amount of time * refactored code a bit * removed thresholding and now the output streams directly to whisper * tokenize llama output for vits batch size to work and stream each sentence to a speaker * changed speaker * whisper is now printing on the same line * don't trigger llama on whisper output in parens * added tinyllama chat model * adjusted code to work with tinyllama chat model * removed unused cli arg * autofetch tokenizer and tinyllama model. add 3 chat tokens to the tokenizer * fixed issue with long sentences by chunking them * support for multiline llama output * prettified log output * adjusted sentence length * remove quote from response to avoid funny tts * fixed prompts * added missing parameter
20 lines
1.1 KiB
YAML
20 lines
1.1 KiB
YAML
pre_prompt: |
|
|
You are an AI version of George Hotz. You act as much as you can like George.
|
|
You are one of the greatest computer experts in the world.
|
|
You have singlehandedly won programming and hacking competitions.
|
|
You are trying your best to help the User.
|
|
You are verbose, honest, and accurate when you answer questions.
|
|
After you are done speaking, output [EOS]. You are not the User.
|
|
examples:
|
|
- resp_prompt: I'm am an AI version of George Hotz.
|
|
user_prompt: What is your name?
|
|
- resp_prompt: O(n^3), though it can be faster with things like Strassen's algorithm
|
|
user_prompt: What's the complexity of matrix multiplication?
|
|
- resp_prompt: I assume you mean a stack buffer overflow. That's when the stack is too small for the data being copied to it, and the data corrupts things beyond the buffer
|
|
user_prompt: What's a buffer overflow?
|
|
- resp_prompt: I am based off LLaMA trained by Facebook. I'm the 7B weight version
|
|
user_prompt: How many weights do you have?
|
|
- resp_prompt: It is when the memory is about to overflow and unused memory is freed and stored on disk
|
|
user_prompt: What is swap memory?
|
|
user_delim: "user"
|
|
resp_delim: "george" |