Compare commits

...

2 Commits

Author SHA1 Message Date
Kwindla Hultman Kramer
b20687e32a workflow_test working except for text_input node 2024-11-01 21:56:30 -07:00
hyypeman
388b3a239b hackathon demo 2024-10-21 22:35:20 -07:00
6 changed files with 225 additions and 0 deletions

1
src/pipecat/workflow/.gitignore vendored Normal file
View File

@@ -0,0 +1 @@
*.json

View File

@@ -0,0 +1 @@
python -m pipecat.workflow.workflow_test to run

View File

View File

@@ -0,0 +1,18 @@
from ..services.cartesia import CartesiaTTSService
from ..services.openai import OpenAILLMService
from ..services.deepgram import DeepgramSTTService
from ..transports.services.daily import DailyTransport
from ..processors.frame_processor import FrameProcessor
# Map workflow types to their corresponding Python classes
WORKFLOW_MAPPING = {
"inputs/audio_input": DailyTransport,
"processors/speech_to_text": DeepgramSTTService,
"processors/llm": OpenAILLMService,
"processors/text_to_speech": CartesiaTTSService,
"outputs/audio_output": DailyTransport,
}
def get_processor_class(node_type: str) -> type[FrameProcessor]:
return WORKFLOW_MAPPING.get(node_type, FrameProcessor)

View File

@@ -0,0 +1,65 @@
import asyncio
import os
from dotenv import load_dotenv
from ..pipeline.pipeline import Pipeline
from ..pipeline.runner import PipelineRunner
from ..pipeline.task import PipelineTask, PipelineParams
from .workflow_translator import translate_workflow
from ..services.openai import OpenAIUserContextAggregator
load_dotenv(override=True)
async def main():
print("Starting workflow test")
# Update the path to the workflow.json file
script_dir = os.path.dirname(os.path.abspath(__file__))
workflow_path = os.path.join(script_dir, "workflow.json")
print(f"Workflow path: {workflow_path}")
# Translate the workflow to a list of processors
print("Translating workflow to processors")
processors, daily_transport = translate_workflow(workflow_path)
print(f"Processors created: {processors}")
# Create a pipeline from the processors
print("Creating pipeline")
pipeline = Pipeline(processors)
print(f"Pipeline created: {pipeline}")
# Create a pipeline task
print("Creating pipeline task")
task = PipelineTask(pipeline, PipelineParams(allow_interruptions=True))
print(f"Pipeline task created: {task}")
# Create a pipeline runner
print("Creating pipeline runner")
runner = PipelineRunner()
print(f"Pipeline runner created: {runner}")
user_context_aggregator = next(
p for p in processors if isinstance(p, OpenAIUserContextAggregator)
)
@daily_transport.event_handler("on_first_participant_joined")
async def on_first_participant_joined(transport, participant):
transport.capture_participant_transcription(participant["id"])
await task.queue_frames([user_context_aggregator.get_context_frame()])
# Run the pipeline
print("Running the pipeline")
try:
await runner.run(task)
print("Pipeline execution completed successfully")
except Exception as e:
print(f"Error during pipeline execution: {e}")
print("Workflow test completed")
if __name__ == "__main__":
print("Starting main execution")
asyncio.run(main())
print("Main execution completed")

View File

@@ -0,0 +1,140 @@
import json
import os
from typing import Any, Dict, List, Tuple
from .workflow_mapping import get_processor_class
from ..processors.frame_processor import FrameProcessor
from ..transports.services.daily import DailyParams
from ..processors.aggregators.openai_llm_context import OpenAILLMContext
from ..audio.vad.silero import SileroVADAnalyzer
from ..transports.base_transport import BaseTransport
def load_workflow(file_path: str) -> Dict[str, Any]:
print(f"Loading workflow from file: {file_path}")
try:
with open(file_path, "r") as f:
workflow = json.load(f)
print(f"Workflow loaded successfully: {workflow}")
return workflow
except Exception as e:
print(f"Error loading workflow: {e}")
raise
def create_processor(node: Dict[str, Any], next_node: Dict[str, Any] = None) -> FrameProcessor:
print(f"Creating processor for node: {node['id']} of type: {node['type']}")
processor_class = get_processor_class(node["type"])
print(f"Processor class: {processor_class}")
# Extract relevant properties for initialization
init_params = {}
if node["type"] == "inputs/audio_input":
init_params = {
"room_url": os.getenv("DAILY_SAMPLE_ROOM_URL"),
"token": "",
"bot_name": "PipecatBot",
"params": DailyParams(
audio_out_enabled=True,
vad_enabled=True,
vad_audio_passthrough=True,
vad_analyzer=SileroVADAnalyzer(),
),
}
elif node["type"] == "processors/speech_to_text":
init_params = {
"api_key": os.getenv("DEEPGRAM_API_KEY"),
}
elif node["type"] == "processors/text_to_speech":
init_params = {
"api_key": os.getenv("CARTESIA_API_KEY"),
"voice_id": "79a125e8-cd45-4c13-8a67-188112f4dd22",
}
print(f"Initialization parameters: {init_params}")
processor = processor_class(**init_params)
print(f"Processor created: {processor}")
return processor
def create_pipeline(workflow: Dict[str, Any]) -> Tuple[List[FrameProcessor], BaseTransport]:
print("Creating pipeline from workflow")
nodes = {node["id"]: node for node in workflow["nodes"]}
links = workflow["links"]
print(f"Nodes: {nodes}")
print(f"Links: {links}")
# Create a dictionary to store processors
processors = {}
daily_transport = None
llm_service = None
context_aggregator = None
# Create processors for each node
for node_id, node in nodes.items():
print(f"Creating processor for node: {node_id}")
if node["type"] == "inputs/audio_input":
daily_transport = create_processor(node)
processors[node_id] = {"processor": daily_transport, "type": node["type"]}
elif node["type"] == "outputs/audio_output":
if daily_transport is None:
raise ValueError("Audio output transport node found before audio input node")
processors[node_id] = {"processor": daily_transport, "type": node["type"]}
elif node["type"] == "processors/llm":
llm_service = create_processor(node)
processors[node_id] = {"processor": llm_service, "type": node["type"]}
context = OpenAILLMContext(
[
{
"role": "system",
"content": "You are a helpful assistant. Your name is Housecat. You are participating in a voice conversation. Keep your answers brief. For punctuation use only period, comma, and question mark.",
},
{"role": "user", "content": "Introduce yourself."},
]
)
context_aggregator = llm_service.create_context_aggregator(context)
print(f"Context aggregator created: {context_aggregator}")
else:
processors[node_id] = {"processor": create_processor(node), "type": node["type"]}
# Create the pipeline based on the links
pipeline = []
for link in links:
source_id, _, _, target_id, _, _ = link
print(f"Processing link: {source_id} -> {target_id}")
if processors[source_id]["processor"] not in pipeline:
print(f"Adding source processor: {source_id}, {processors[source_id]['processor']}")
if processors[source_id]["type"] == "inputs/audio_input":
pipeline.append(processors[source_id]["processor"].input())
else:
pipeline.append(processors[source_id]["processor"])
if processors[target_id]["processor"] not in pipeline and target_id in processors:
print(f"Adding target processor: {target_id} {processors[target_id]['processor']}")
if processors[target_id]["type"] == "outputs/audio_output":
pipeline.append(processors[target_id]["processor"].output())
elif processors[target_id]["type"] == "processors/llm":
print("TRYING TO LINK AGGREGATOR")
if context_aggregator:
print("AGGREGATOR FOUND")
pipeline.append(context_aggregator.user())
pipeline.append(processors[target_id]["processor"])
else:
pipeline.append(processors[target_id]["processor"])
print(f"Pipeline created with {len(pipeline)} processors")
print(f"Pipeline: {pipeline}")
return pipeline, daily_transport
def translate_workflow(file_path: str) -> Tuple[List[FrameProcessor], BaseTransport]:
print(f"Translating workflow from file: {file_path}")
workflow = load_workflow(file_path)
pipeline, transport = create_pipeline(workflow)
print("Workflow translation completed")
return pipeline, transport