move llm config to profiles directory

software/source/clients/base_device.py

@@ -364,6 +364,10 @@ class Device:
                 if type(chunk) == str:
                     chunk = json.loads(chunk)
 
+                    if chunk.get("type") == "config":
+                        self.tts_service = chunk.get("tts_service")
+                        continue
+
                 if self.tts_service == "elevenlabs":
                     message = chunk
                 else:

software/source/server/async_interpreter.py

@@ -194,7 +194,7 @@ class AsyncInterpreter:
 
         self.tts.feed(text_iterator)
 
-        self.tts.play_async(on_audio_chunk=self.on_tts_chunk, muted=False)
+        self.tts.play_async(on_audio_chunk=self.on_tts_chunk, muted=True)
 
         while True:
             await asyncio.sleep(0.1)

software/source/server/async_server.py

@@ -1,10 +1,9 @@
-# make this obvious
+# TODO: import from the profiles directory the interpreter directory
-from .profiles.default import interpreter as base_interpreter
+
+from .profiles.fast import interpreter as base_interpreter
 
-# from .profiles.fast import interpreter as base_interpreter
 # from .profiles.local import interpreter as base_interpreter
-
+# from .profiles.default import interpreter as base_interpreter
-# TODO: remove files i.py, llm.py, conftest?, services
 
 import asyncio
 import traceback
@@ -12,8 +11,6 @@ import json
 from fastapi import FastAPI, WebSocket
 from fastapi.responses import PlainTextResponse
 from uvicorn import Config, Server
-
-# from interpreter import interpreter as base_interpreter
 from .async_interpreter import AsyncInterpreter
 from fastapi.middleware.cors import CORSMiddleware
 from typing import List, Dict, Any
@@ -24,8 +21,8 @@ os.environ["STT_RUNNER"] = "server"
 os.environ["TTS_RUNNER"] = "server"
 
 
-async def main(server_host, server_port, tts_service):
+async def main(server_host, server_port):
-    base_interpreter.tts = tts_service
+    # interpreter.tts set in the profiles directory!!!!
     interpreter = AsyncInterpreter(base_interpreter)
 
     app = FastAPI()
@@ -52,6 +49,12 @@ async def main(server_host, server_port, tts_service):
     @app.websocket("/")
     async def websocket_endpoint(websocket: WebSocket):
         await websocket.accept()
+
+        # Send the tts_service value to the client
+        await websocket.send_text(
+            json.dumps({"type": "config", "tts_service": interpreter.interpreter.tts})
+        )
+
         try:
 
             async def receive_input():

software/source/server/profiles/default.py

@@ -3,6 +3,9 @@ from interpreter import interpreter
 # This is an Open Interpreter compatible profile.
 # Visit https://01.openinterpreter.com/profile for all options.
 
+# 01 suports OpenAI, ElevenLabs, and Coqui (Local) TTS providers
+# {OpenAI: "openai", ElevenLabs: "elevenlabs", Coqui: "coqui"}
+interpreter.tts = "openai"
 
 # Connect your 01 to a language model
 interpreter.llm.model = "gpt-4-turbo"
@@ -10,9 +13,6 @@ interpreter.llm.context_window = 100000
 interpreter.llm.max_tokens = 4096
 # interpreter.llm.api_key = "<your_openai_api_key_here>"
 
-# Give your 01 a voice
-interpreter.tts = "openai"
-
 # Tell your 01 where to find and save skills
 interpreter.computer.skills.path = "./skills"
 

software/source/server/profiles/fast.py

@@ -1,31 +1,42 @@
-async def main(server_host, server_port, tts_service, asynchronous):
+from interpreter import interpreter
 
-    if asynchronous:
+# This is an Open Interpreter compatible profile.
+# Visit https://01.openinterpreter.com/profile for all options.
 
-        base_interpreter.system_message = (
+# 01 suports OpenAI, ElevenLabs, and Coqui (Local) TTS providers
+# {OpenAI: "openai", ElevenLabs: "elevenlabs", Coqui: "coqui"}
+interpreter.tts = "elevenlabs"
+
+# 01 Language Model Config.
+interpreter.llm_service = "litellm"
+interpreter.llm.model = "groq/llama3-8b-8192"
+interpreter.llm.supports_vision = False
+interpreter.llm.supports_functions = False
+interpreter.llm.context_window = 2048
+interpreter.llm.max_tokens = 4096
+interpreter.llm.temperature = 0.8
+
+# interpreter.llm.api_key = os.environ["GROQ_API_KEY"]
+
+interpreter.computer.import_computer_api = False
+
+interpreter.auto_run = True
+interpreter.system_message = (
     "You are a helpful assistant that can answer questions and help with tasks."
 )
 
-        base_interpreter.computer.import_computer_api = False
+# TODO: include other options in comments in the profiles for tts
+# direct people to the profiles directory to make changes to the interpreter profile
+# this should be made explicit on the docs
 
-        base_interpreter.llm.model = "groq/llama3-8b-8192"
+"""
-
+    llm_service: str = "litellm",
-        base_interpreter.llm.api_key = os.environ["GROQ_API_KEY"]
+    model: str = "gpt-4",
-
+    llm_supports_vision: bool = False,
-        base_interpreter.llm.supports_functions = False
+    llm_supports_functions: bool = False,
-
+    context_window: int = 2048,
-        base_interpreter.auto_run = True
+    max_tokens: int = 4096,
-
+    temperature: float = 0.8,
-        base_interpreter.tts = tts_service
+    tts_service: str = "elevenlabs",
-
+    stt_service: str = "openai",
-        interpreter = AsyncInterpreter(base_interpreter)
+"""
-
-    else:
-
-        configured_interpreter = configure_interpreter(base_interpreter)
-
-        configured_interpreter.llm.supports_functions = True
-
-        configured_interpreter.tts = tts_service
-
-        interpreter = AsyncInterpreter(configured_interpreter)

software/source/server/profiles/local.py

@@ -1,5 +1,9 @@
 from interpreter import interpreter
 
+# 01 suports OpenAI, ElevenLabs, and Coqui (Local) TTS providers
+# {OpenAI: "openai", ElevenLabs: "elevenlabs", Coqui: "coqui"}
+interpreter.tts = "coqui"
+
 # Local setup
 interpreter.local_setup()