init

.gitignore

@@ -0,0 +1 @@
+config.py

README.md

@@ -0,0 +1,24 @@
+
+# AI Waifu Vtuber & Assistant
+
+This project is inspired by shioridotdev and utilizes various technologies such as VoiceVox Engine, DeepL, Whisper OpenAI, and VtubeStudio to create an AI waifu virtual YouTuber.
+
+
+
+
+## Technologies Used
+
+ - VoiceVox Engine
+ - DeepL
+ - Whisper OpenAI
+ - VtubeStudio
+
+
+## Installation
+
+To use this project, you will need to download the VoiceVox Engine from their GitHub page. You will also need to capture desktop audio using Virtual Cable and use it as input on VtubeStudio microphone.
+
+## Credits
+
+This project is inspired by the work of shioridotdev. Special thanks to the creators of the technologies used in this project including VoiceVox Engine, DeepL, Whisper OpenAI, and VtubeStudio.
+

run.py

@@ -0,0 +1,147 @@
+import deepl
+import urllib
+import urllib.parse
+import urllib.request
+import requests
+import openai
+import winsound
+import sys
+import pytchat
+import time
+import re
+import pyaudio
+import keyboard
+import wave
+from config import *
+
+sys.stdout = open(sys.stdout.fileno(), mode='w', encoding='utf8', buffering=1)
+
+openai.api_key = api_key #API From OpenAI
+translator = deepl.Translator(auth_key) #API from DeepL
+
+conversation = [{"role": "system", "content": "you are an AI Waifu Vtuber called Pina. You reply with brief, to-the-point answers with no elaboration."}]
+total_characters = 0
+chat_raw = ""
+
+def record_audio():
+    CHUNK = 1024
+    FORMAT = pyaudio.paInt16
+    CHANNELS = 1
+    RATE = 44100
+    RECORD_SECONDS = 5
+    WAVE_OUTPUT_FILENAME = "input.wav"
+    p = pyaudio.PyAudio()
+    stream = p.open(format=FORMAT,
+                    channels=CHANNELS,
+                    rate=RATE,
+                    input=True,
+                    frames_per_buffer=CHUNK)
+    frames = []
+    print("Recording...")
+    while keyboard.is_pressed('RIGHT_SHIFT'):
+        data = stream.read(CHUNK)
+        frames.append(data)
+    print("Stopped recording.")
+    stream.stop_stream()
+    stream.close()
+    p.terminate()
+    wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
+    wf.setnchannels(CHANNELS)
+    wf.setsampwidth(p.get_sample_size(FORMAT))
+    wf.setframerate(RATE)
+    wf.writeframes(b''.join(frames))
+    wf.close()
+    transcribe_audio("input.wav")
+
+def transcribe_audio(file):
+    try:
+        audio_file= open(file, "rb")
+        transcript = openai.Audio.translate("whisper-1", audio_file)
+    except:
+        print("Error transcribing audio")
+    chat_raw = transcript.text
+    openai_answer(chat_raw)
+    # translate_text(transcript.text)
+
+def openai_answer(chat):
+    result = translator.translate_text(chat, target_lang="EN-US")
+    print ("Question: " + result.text)
+    conversation.append({"role": "user", "content": result.text})
+    total_characters = sum(len(d['content']) for d in conversation)
+
+    while total_characters > 4000 and len(conversation) > 1:
+        # remove the second dictionary from the list
+        conversation.pop(1)
+
+    response = openai.ChatCompletion.create(
+        model="gpt-3.5-turbo",
+        messages=conversation,
+        max_tokens=500,
+        temperature=1,
+        top_p=0.9
+    )
+    conversation.append({"role": "assistant", "content": response['choices'][0]['message']['content']})
+    message = response['choices'][0]['message']['content']
+    translate_text(message)
+
+def translate_text(text):
+    result_id = translator.translate_text(text, target_lang="ID")
+    print("ID Answer: " + result_id.text)
+    result_jp = translator.translate_text(text, target_lang="JA")
+    print("JP Answer: " + result_jp.text)
+    speech_text(result_jp.text, result_id.text)
+
+def speech_text(result_jp, result_id):
+    # Install and run 
+    # http://127.0.0.1:50021/docs API Docs
+    params_encoded = urllib.parse.urlencode({'text': result_jp, 'speaker': 46})
+    request = requests.post(f'http://127.0.0.1:50021/audio_query?{params_encoded}')
+    params_encoded = urllib.parse.urlencode({'speaker': 46, 'enable_interrogative_upspeak': True})
+    request = requests.post(f'http://127.0.0.1:50021/synthesis?{params_encoded}', json=request.json())
+
+    with open("output.wav", "wb") as outfile:
+        outfile.write(request.content)
+
+    with open("output.txt", "w") as outfile:
+        text = result_id
+        text = text.replace("?", "?\n")
+        text = text.replace("!", "!\n")
+        text = text.replace(".", ".\n")
+        outfile.write(text)
+
+    with open("chat.txt", "w") as outfile:
+        outfile.write(chat_raw)
+
+    winsound.PlaySound("output.wav", winsound.SND_FILENAME)
+
+    time.sleep(1)
+    with open ("output.txt", "w") as f:
+        f.truncate(0)
+    with open ("chat.txt", "w") as f:
+        f.truncate(0)
+
+if __name__ == "__main__":
+
+    mode = input("Mode (1-Mic, 2-Youtube Live): ")
+
+    if mode == "1":
+        print("Press Right Shift to record audio")
+        while True:
+            if keyboard.is_pressed('RIGHT_SHIFT'):
+                record_audio()
+
+    elif mode == "2":
+        live_id = input("Livestream ID: ")
+        live = pytchat.create(video_id=live_id)
+        while live.is_alive():
+            for c in live.get().sync_items():
+                if c.author.isChatOwner or c.author.name == 'Nightbot':
+                    continue
+                chat_raw = c.author.name + ': ' + c.message
+                chat = re.sub(r':[^\s]+:', '', c.message)
+                if len(chat) > 5:
+                    chat = c.author.name + ' berkata: ' + chat
+                    print(chat)
+                    openai_answer(chat)
+
+