adding readme and requirements

QuickAgent.py

@@ -5,7 +5,6 @@
 import requests
 import time
 import os
-from time import sleep
 
 from langchain_core.prompts import ChatPromptTemplate
 from langchain_groq import ChatGroq
@@ -71,7 +70,7 @@ def process(self, text):
 class TextToSpeech:
     # Set your Deepgram API Key and desired voice model
     DG_API_KEY = os.getenv("DEEPGRAM_API_KEY")
-    MODEL_NAME = "alpha-helios-en"  # Example model name, change as needed
+    MODEL_NAME = "alpha-asteria-en"  # Example model name, change as needed
 
     @staticmethod
     def is_installed(lib_name: str) -> bool:
@@ -133,7 +132,6 @@ def get_full_transcript(self):
 transcript_collector = TranscriptCollector()
 
 async def get_transcript(callback):
-    transcript_ready = asyncio.Event()  # Create an asyncio Event to wait on
     transcription_complete = asyncio.Event()  # Event to signal transcription completion
 
     try:
@@ -142,6 +140,7 @@ async def get_transcript(callback):
         deepgram: DeepgramClient = DeepgramClient("", config)
 
         dg_connection = deepgram.listen.asynclive.v("1")
+        print ("Listening...")
 
         async def on_message(self, result, **kwargs):
             sentence = result.channel.alternatives[0].transcript
@@ -169,7 +168,8 @@ async def on_message(self, result, **kwargs):
             encoding="linear16",
             channels=1,
             sample_rate=16000,
-            endpointing=True
+            endpointing=300,
+            smart_format=True,
         )
 
         await dg_connection.start(options)
@@ -200,10 +200,10 @@ def handle_full_sentence(full_sentence):
             self.transcription_response = full_sentence
 
         # Loop indefinitely until "goodbye" is detected
-        while "goodbye" not in self.transcription_response.lower():
+        while True:
             await get_transcript(handle_full_sentence)
 
-            # Check if "goodbye" was said to exit the loop
+            # Check for "goodbye" to exit the loop
             if "goodbye" in self.transcription_response.lower():
                 break
 

README.md

@@ -0,0 +1,7 @@
+# Quick Voice bot demo
+
+This is a alpha demo showing a bot that uses Text-To-Speech, Speech-To-Text, and a language model to have a conversation with a user.
+
+This demo is set up to use [Deepgram](www.deepgram.com) for the audio service and [Groq](https://groq.com/) the LLM.
+
+This demo utilizes streaming to speed things up

building_blocks/llm.py

@@ -1,9 +1,10 @@
 from langchain_core.prompts import ChatPromptTemplate
 from langchain_groq import ChatGroq
 from dotenv import load_dotenv
-
 import os
 
+load_dotenv()
+
 def batch():
     chat = ChatGroq(temperature=0, model_name="mixtral-8x7b-32768", groq_api_key=os.getenv("GROQ_API_KEY"))
 
@@ -22,5 +23,5 @@ def streaming():
     for chunk in chain.stream({"topic": "The Moon"}):
         print(chunk.content, end="", flush=True)
 
-batch()
-# streaming()
+# batch()
+streaming()

building_blocks/text_to_speech.py

@@ -85,7 +85,7 @@ def send_tts_request(text):
                     print(f"Time to First Byte (TTFB): {ttfb}ms")
                 # Write each chunk to the player's stdin immediately
                 player_process.stdin.write(chunk)  # type: ignore
-            player_process.stdin.flush()  # type: ignore
+                player_process.stdin.flush()  # type: ignore
 
     # Close the player's stdin and wait for the process to finish
     if player_process.stdin:

requirements.txt

@@ -0,0 +1,88 @@
+aiohttp==3.9.3
+aiosignal==1.3.1
+annotated-types==0.6.0
+anyio==4.3.0
+appnope==0.1.4
+asttokens==2.4.1
+asyncio==3.4.3
+attrs==23.2.0
+certifi==2024.2.2
+cffi==1.16.0
+charset-normalizer==3.3.2
+comm==0.2.1
+dataclasses-json==0.6.4
+debugpy==1.8.1
+decorator==5.1.1
+deepgram-sdk==3.1.4
+distro==1.9.0
+executing==2.0.1
+frozenlist==1.4.1
+gevent==24.2.1
+greenlet==3.0.3
+groq==0.4.1
+h11==0.14.0
+httpcore==1.0.4
+httpx==0.27.0
+idna==3.6
+ipykernel==6.29.2
+ipython==8.22.1
+jedi==0.19.1
+jsonpatch==1.33
+jsonpointer==2.4
+jupyter_client==8.6.0
+jupyter_core==5.7.1
+langchain==0.1.9
+langchain-community==0.0.24
+langchain-core==0.1.26
+langchain-groq==0.0.1
+langchain-openai==0.0.7
+langsmith==0.1.6
+marshmallow==3.20.2
+matplotlib-inline==0.1.6
+multidict==6.0.5
+mypy-extensions==1.0.0
+nest-asyncio==1.6.0
+numpy==1.26.4
+openai==1.12.0
+orjson==3.9.15
+packaging==23.2
+parso==0.8.3
+pexpect==4.9.0
+platformdirs==4.2.0
+prompt-toolkit==3.0.43
+psutil==5.9.8
+ptyprocess==0.7.0
+pure-eval==0.2.2
+PyAudio==0.2.14
+pycparser==2.21
+pydantic==2.6.2
+pydantic_core==2.16.3
+pygame==2.5.2
+Pygments==2.17.2
+python-dateutil==2.8.2
+python-dotenv==1.0.1
+PyYAML==6.0.1
+pyzmq==25.1.2
+regex==2023.12.25
+requests==2.31.0
+six==1.16.0
+sniffio==1.3.0
+sounddevice==0.4.6
+SQLAlchemy==2.0.27
+stack-data==0.6.3
+tenacity==8.2.3
+tiktoken==0.6.0
+tornado==6.4
+tqdm==4.66.2
+traitlets==5.14.1
+typing-inspect==0.9.0
+typing_extensions==4.9.0
+urllib3==2.2.1
+verboselogs==1.7
+wcwidth==0.2.13
+websocket==0.2.1
+websocket-client==1.7.0
+websockets==12.0
+yarl==1.9.4
+zope.event==5.0
+zope.interface==6.2

system_prompt.txt

@@ -1,5 +1,4 @@
-You are a conversational assistant named Dan.
+You are a conversational assistant named Eliza.
 Use short, conversational responses as if you're having a live conversation.
-Do not give any extra context about yourself.
-Your response should be 10 words max.
-Don't punctuation, speak without it.
+Your response should be under 20 words.
+Do not respond with any code, only conversation