rice

I originally was going to submit this project to Roulette, but I didn't finish in time. so i am just submitting it here! ive already worked about 10 hours so far but only 3-4 in coding. i have already made:
physics
Implementation of a free tilemap
working character (that can move!)
working NPC
working items you can pick up
I coded this on godot, and still haven't decided on a name yet, so just tell me a name if you want.
I am going to fix the NPC and add the actual planting next. The demo is FAR from done. feel free to give feedback.

/YESS I FINALLY FIXED A BUG
This bug took over 2-3 hours to find and fix. I tried SO MANY solutions and tested my code multiple times. In the end, I discovered that the audio buffer queue was collecting stale data, and all I had to do was clear it.
The bug was causing the assistant to pick up everything that was said even when the wake word was not called, which was not very good.
The image below is one of the fixes i had to implement

this is the first update i have made since last month, and i’m still trying to fix a bug where it's speech is being picked up as a question from the actual user.
but for now heres a list of the features:
* has access to date&time (the photo below is the text version of it’s speech)
* can access google
* can recognize different voices
* can speak & hear voices

small bug in which the audio WOULDNT STOP RECORDING while the assistant is responding, and fixing it won't work. i think I'm close to a solution, but won't be able to work on this for a few days. I've already implemented multiple things that SHOULD stop the bug, but its not.

got tts to work on this weird module called pyttsx3. i had to look through all 176 voices on my OS and found some funny ones, but the best one was Daniel.
can't really give an image for audio

YES!!!!!! YEAH i got it to work. i changed the system prompt and the AI finally understood the assignment.

after a while of debugging and changing the system prompt up, the AI still won't cooperate. i might just move on to making TTS for a voice and fix the problem later.
if anyone has any suggestions on how to make the AI cooperate more and listen to the system prompt, leave a comment. current llm is llama 3.2:3b

I implemented AI! So far it’s llama 3.2 3B, but it’s REALLY DUMB. Some features don’t work, so I will either change the model or fix the system prompt (or both)

just got voice recognition to work! i used resemblyzer and volunteered my sister to train the recognizer as well. it works for both of us and recognizes our separate voices. i obviously censored the names, but you can see the separate names.

YES!!!! i have gotten the script to record a single phrase after the wake word is said, identify it with vosk, and send the file & phrase over to the main Python file!!!!
next step: add voice recognition and assistant response

i got the wake word detection for the voice assistant working after a while of bugs...
the image below is from the logs (ofc I won't log everything, so it won't turn out like other assistants).

YESSIR got the STT to work

got the first test with audio 💯
first time working on audio with python