Rename services/ to voice/ and function-calling/, flatten to top level

Replace the nested services/speech/ and services/function-calling/ with
top-level voice/ and function-calling/ directories. Update eval script
paths and README to match.
This commit is contained in:
Mark Backman
2026-03-31 15:12:06 -04:00
parent f14638a1fd
commit 47b41a0ff7
93 changed files with 96 additions and 260 deletions

View File

@@ -61,12 +61,13 @@ uv run getting-started/06-voice-agent.py -t twilio -x NGROK_HOST_NAME
Progressive introduction to Pipecat, from minimal TTS to a full voice agent with function calling.
### [`services/`](./services/)
### [`voice/`](./voice/)
Service provider integration examples, organized into subfolders:
Full STT + LLM + TTS voice agent pipelines showcasing different speech service providers (Deepgram, ElevenLabs, Cartesia, etc.)
- **[`speech/`](./services/speech/)** — Full STT + LLM + TTS pipelines showcasing different speech service providers (Deepgram, ElevenLabs, Cartesia, etc.)
- **[`function-calling/`](./services/function-calling/)** — Function calling with different LLM providers (OpenAI, Anthropic, Google, etc.)
### [`function-calling/`](./function-calling/)
Function calling with different LLM providers (OpenAI, Anthropic, Google, etc.)
### [`transcription/`](./transcription/)

View File

@@ -1,165 +0,0 @@
#
# Copyright (c) 2024-2026, Daily
#
# SPDX-License-Identifier: BSD 2-Clause License
#
import os
from dotenv import load_dotenv
from loguru import logger
from pipecat.adapters.schemas.function_schema import FunctionSchema
from pipecat.adapters.schemas.tools_schema import ToolsSchema
from pipecat.audio.vad.silero import SileroVADAnalyzer
from pipecat.frames.frames import LLMRunFrame, TTSSpeakFrame
from pipecat.pipeline.pipeline import Pipeline
from pipecat.pipeline.runner import PipelineRunner
from pipecat.pipeline.task import PipelineParams, PipelineTask
from pipecat.processors.aggregators.llm_context import LLMContext
from pipecat.processors.aggregators.llm_response_universal import (
LLMContextAggregatorPair,
LLMUserAggregatorParams,
)
from pipecat.runner.types import RunnerArguments
from pipecat.runner.utils import create_transport
from pipecat.services.deepgram.stt import DeepgramSTTService
from pipecat.services.elevenlabs.tts import ElevenLabsTTSService
from pipecat.services.google.vertex.llm import GoogleVertexLLMService
from pipecat.services.llm_service import FunctionCallParams
from pipecat.transports.base_transport import BaseTransport, TransportParams
from pipecat.transports.daily.transport import DailyParams
from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
load_dotenv(override=True)
async def fetch_weather_from_api(params: FunctionCallParams):
await params.result_callback({"conditions": "nice", "temperature": "75"})
# We use lambdas to defer transport parameter creation until the transport
# type is selected at runtime.
transport_params = {
"daily": lambda: DailyParams(
audio_in_enabled=True,
audio_out_enabled=True,
),
"twilio": lambda: FastAPIWebsocketParams(
audio_in_enabled=True,
audio_out_enabled=True,
),
"webrtc": lambda: TransportParams(
audio_in_enabled=True,
audio_out_enabled=True,
),
}
async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
logger.info(f"Starting bot")
stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
tts = ElevenLabsTTSService(
api_key=os.getenv("ELEVENLABS_API_KEY", ""),
settings=ElevenLabsTTSService.Settings(
voice=os.getenv("ELEVENLABS_VOICE_ID", ""),
),
)
llm = GoogleVertexLLMService(
credentials=os.getenv("GOOGLE_VERTEX_TEST_CREDENTIALS"),
project_id=os.getenv("GOOGLE_CLOUD_PROJECT_ID"),
location=os.getenv("GOOGLE_CLOUD_LOCATION"),
settings=GoogleVertexLLMService.Settings(
system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
),
)
# You can aslo register a function_name of None to get all functions
# sent to the same callback with an additional function_name parameter.
llm.register_function("get_current_weather", fetch_weather_from_api)
@llm.event_handler("on_function_calls_started")
async def on_function_calls_started(service, function_calls):
await tts.queue_frame(TTSSpeakFrame("Let me check on that."))
weather_function = FunctionSchema(
name="get_current_weather",
description="Get the current weather",
properties={
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA",
},
"format": {
"type": "string",
"enum": ["celsius", "fahrenheit"],
"description": "The temperature unit to use. Infer this from the user's location.",
},
},
required=["location", "format"],
)
tools = ToolsSchema(standard_tools=[weather_function])
messages = [
{
"role": "developer",
"content": "Start a conversation with 'Hey there' to get the current weather.",
},
]
context = LLMContext(messages, tools)
user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
context,
user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
)
pipeline = Pipeline(
[
transport.input(),
stt,
user_aggregator,
llm,
tts,
transport.output(),
assistant_aggregator,
]
)
task = PipelineTask(
pipeline,
params=PipelineParams(
enable_metrics=True,
enable_usage_metrics=True,
),
idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
)
@transport.event_handler("on_client_connected")
async def on_client_connected(transport, client):
logger.info(f"Client connected")
# Kick off the conversation.
await task.queue_frames([LLMRunFrame()])
@transport.event_handler("on_client_disconnected")
async def on_client_disconnected(transport, client):
logger.info(f"Client disconnected")
await task.cancel()
runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
await runner.run(task)
async def bot(runner_args: RunnerArguments):
"""Main bot entry point compatible with Pipecat Cloud."""
transport = await create_transport(runner_args, transport_params)
await run_bot(transport, runner_args)
if __name__ == "__main__":
from pipecat.runner.run import main
main()

View File

@@ -96,61 +96,61 @@ EVAL_COMPLETE_TURN = EvalConfig(
)
TESTS_SPEECH = [
("services/speech/cartesia.py", EVAL_SIMPLE_MATH),
("services/speech/cartesia-http.py", EVAL_SIMPLE_MATH),
("services/speech/speechmatics.py", EVAL_SIMPLE_MATH),
("services/speech/speechmatics-vad.py", EVAL_SIMPLE_MATH),
("services/speech/langchain.py", EVAL_SIMPLE_MATH),
("services/speech/deepgram.py", EVAL_SIMPLE_MATH),
("services/speech/deepgram-flux.py", EVAL_SIMPLE_MATH),
("services/speech/deepgram-http.py", EVAL_SIMPLE_MATH),
("services/speech/elevenlabs.py", EVAL_SIMPLE_MATH),
("services/speech/elevenlabs-http.py", EVAL_SIMPLE_MATH),
("services/speech/xai.py", EVAL_SIMPLE_MATH),
("services/speech/azure.py", EVAL_SIMPLE_MATH),
("services/speech/azure-http.py", EVAL_SIMPLE_MATH),
("services/speech/openai.py", EVAL_SIMPLE_MATH),
("services/speech/openai-http.py", EVAL_SIMPLE_MATH),
("services/speech/gladia.py", EVAL_SIMPLE_MATH),
("services/speech/gladia-vad.py", EVAL_SIMPLE_MATH),
("services/speech/lmnt.py", EVAL_SIMPLE_MATH),
("services/speech/groq.py", EVAL_SIMPLE_MATH),
("services/speech/aws.py", EVAL_SIMPLE_MATH),
("services/speech/aws-strands.py", EVAL_WEATHER),
("services/speech/google-gemini-tts.py", EVAL_SIMPLE_MATH),
("services/speech/google.py", EVAL_SIMPLE_MATH),
("services/speech/google-http.py", EVAL_SIMPLE_MATH),
("services/speech/assemblyai.py", EVAL_SIMPLE_MATH),
("services/speech/krisp-viva.py", EVAL_SIMPLE_MATH),
("services/speech/rime.py", EVAL_SIMPLE_MATH),
("services/speech/rime-http.py", EVAL_SIMPLE_MATH),
("services/speech/nvidia.py", EVAL_SIMPLE_MATH),
("services/speech/google-audio-in.py", EVAL_SIMPLE_MATH),
("services/speech/fish.py", EVAL_SIMPLE_MATH),
("services/speech/neuphonic.py", EVAL_SIMPLE_MATH),
("services/speech/neuphonic-http.py", EVAL_SIMPLE_MATH),
("services/speech/fal.py", EVAL_SIMPLE_MATH),
("services/speech/minimax.py", EVAL_SIMPLE_MATH),
("services/speech/sarvam.py", EVAL_SIMPLE_MATH),
("services/speech/sarvam-http.py", EVAL_SIMPLE_MATH),
("services/speech/soniox.py", EVAL_SIMPLE_MATH),
("services/speech/inworld.py", EVAL_SIMPLE_MATH),
("services/speech/inworld-http.py", EVAL_SIMPLE_MATH),
("services/speech/asyncai.py", EVAL_SIMPLE_MATH),
("services/speech/asyncai-http.py", EVAL_SIMPLE_MATH),
("services/speech/aicoustics.py", EVAL_SIMPLE_MATH),
("services/speech/hume.py", EVAL_SIMPLE_MATH),
("services/speech/gradium.py", EVAL_SIMPLE_MATH),
("services/speech/camb.py", EVAL_SIMPLE_MATH),
("services/speech/piper.py", EVAL_SIMPLE_MATH),
("services/speech/kokoro.py", EVAL_SIMPLE_MATH),
("services/speech/resemble.py", EVAL_SIMPLE_MATH),
("services/speech/smallest.py", EVAL_SIMPLE_MATH),
("services/speech/openai-responses.py", EVAL_SIMPLE_MATH),
("services/speech/openai-responses-http.py", EVAL_SIMPLE_MATH),
TESTS_VOICE = [
("voice/cartesia.py", EVAL_SIMPLE_MATH),
("voice/cartesia-http.py", EVAL_SIMPLE_MATH),
("voice/speechmatics.py", EVAL_SIMPLE_MATH),
("voice/speechmatics-vad.py", EVAL_SIMPLE_MATH),
("voice/langchain.py", EVAL_SIMPLE_MATH),
("voice/deepgram.py", EVAL_SIMPLE_MATH),
("voice/deepgram-flux.py", EVAL_SIMPLE_MATH),
("voice/deepgram-http.py", EVAL_SIMPLE_MATH),
("voice/elevenlabs.py", EVAL_SIMPLE_MATH),
("voice/elevenlabs-http.py", EVAL_SIMPLE_MATH),
("voice/xai.py", EVAL_SIMPLE_MATH),
("voice/azure.py", EVAL_SIMPLE_MATH),
("voice/azure-http.py", EVAL_SIMPLE_MATH),
("voice/openai.py", EVAL_SIMPLE_MATH),
("voice/openai-http.py", EVAL_SIMPLE_MATH),
("voice/gladia.py", EVAL_SIMPLE_MATH),
("voice/gladia-vad.py", EVAL_SIMPLE_MATH),
("voice/lmnt.py", EVAL_SIMPLE_MATH),
("voice/groq.py", EVAL_SIMPLE_MATH),
("voice/aws.py", EVAL_SIMPLE_MATH),
("voice/aws-strands.py", EVAL_WEATHER),
("voice/google-gemini-tts.py", EVAL_SIMPLE_MATH),
("voice/google.py", EVAL_SIMPLE_MATH),
("voice/google-http.py", EVAL_SIMPLE_MATH),
("voice/assemblyai.py", EVAL_SIMPLE_MATH),
("voice/krisp-viva.py", EVAL_SIMPLE_MATH),
("voice/rime.py", EVAL_SIMPLE_MATH),
("voice/rime-http.py", EVAL_SIMPLE_MATH),
("voice/nvidia.py", EVAL_SIMPLE_MATH),
("voice/google-audio-in.py", EVAL_SIMPLE_MATH),
("voice/fish.py", EVAL_SIMPLE_MATH),
("voice/neuphonic.py", EVAL_SIMPLE_MATH),
("voice/neuphonic-http.py", EVAL_SIMPLE_MATH),
("voice/fal.py", EVAL_SIMPLE_MATH),
("voice/minimax.py", EVAL_SIMPLE_MATH),
("voice/sarvam.py", EVAL_SIMPLE_MATH),
("voice/sarvam-http.py", EVAL_SIMPLE_MATH),
("voice/soniox.py", EVAL_SIMPLE_MATH),
("voice/inworld.py", EVAL_SIMPLE_MATH),
("voice/inworld-http.py", EVAL_SIMPLE_MATH),
("voice/asyncai.py", EVAL_SIMPLE_MATH),
("voice/asyncai-http.py", EVAL_SIMPLE_MATH),
("voice/aicoustics.py", EVAL_SIMPLE_MATH),
("voice/hume.py", EVAL_SIMPLE_MATH),
("voice/gradium.py", EVAL_SIMPLE_MATH),
("voice/camb.py", EVAL_SIMPLE_MATH),
("voice/piper.py", EVAL_SIMPLE_MATH),
("voice/kokoro.py", EVAL_SIMPLE_MATH),
("voice/resemble.py", EVAL_SIMPLE_MATH),
("voice/smallest.py", EVAL_SIMPLE_MATH),
("voice/openai-responses.py", EVAL_SIMPLE_MATH),
("voice/openai-responses-http.py", EVAL_SIMPLE_MATH),
# Needs a local XTTS docker instance running.
# ("services/speech/xtts.py", EVAL_SIMPLE_MATH),
# ("voice/xtts.py", EVAL_SIMPLE_MATH),
]
TESTS_VISION = [
@@ -169,44 +169,44 @@ TESTS_VISION = [
TESTS_FUNCTION_CALLING = [
("getting-started/07-function-calling.py", EVAL_WEATHER),
("getting-started/07-function-calling.py", EVAL_WEATHER_AND_RESTAURANT),
("services/function-calling/openai-responses.py", EVAL_WEATHER),
("services/function-calling/openai-responses.py", EVAL_WEATHER_AND_RESTAURANT),
("services/function-calling/openai-responses-http.py", EVAL_WEATHER),
("services/function-calling/openai-responses-http.py", EVAL_WEATHER_AND_RESTAURANT),
("services/function-calling/anthropic.py", EVAL_WEATHER),
("services/function-calling/anthropic.py", EVAL_WEATHER_AND_RESTAURANT),
("services/function-calling/openai.py", EVAL_WEATHER),
("services/function-calling/google.py", EVAL_WEATHER),
("services/function-calling/google.py", EVAL_WEATHER_AND_RESTAURANT),
("services/function-calling/groq.py", EVAL_WEATHER),
("services/function-calling/grok.py", EVAL_WEATHER),
("services/function-calling/azure.py", EVAL_WEATHER),
("services/function-calling/fireworks.py", EVAL_WEATHER),
("services/function-calling/nvidia.py", EVAL_WEATHER),
("services/function-calling/cerebras.py", EVAL_WEATHER),
("services/function-calling/openrouter.py", EVAL_WEATHER),
("services/function-calling/perplexity.py", EVAL_WEATHER),
("services/function-calling/google-vertex.py", EVAL_WEATHER),
("services/function-calling/qwen.py", EVAL_WEATHER),
("services/function-calling/aws.py", EVAL_WEATHER),
("services/function-calling/sambanova.py", EVAL_WEATHER),
("services/function-calling/aws.py", EVAL_WEATHER_AND_RESTAURANT),
("services/function-calling/nebius.py", EVAL_WEATHER),
("services/function-calling/mistral.py", EVAL_WEATHER),
("services/function-calling/sarvam.py", EVAL_WEATHER),
("services/function-calling/novita.py", EVAL_WEATHER),
("function-calling/openai-responses.py", EVAL_WEATHER),
("function-calling/openai-responses.py", EVAL_WEATHER_AND_RESTAURANT),
("function-calling/openai-responses-http.py", EVAL_WEATHER),
("function-calling/openai-responses-http.py", EVAL_WEATHER_AND_RESTAURANT),
("function-calling/anthropic.py", EVAL_WEATHER),
("function-calling/anthropic.py", EVAL_WEATHER_AND_RESTAURANT),
("function-calling/openai.py", EVAL_WEATHER),
("function-calling/google.py", EVAL_WEATHER),
("function-calling/google.py", EVAL_WEATHER_AND_RESTAURANT),
("function-calling/groq.py", EVAL_WEATHER),
("function-calling/grok.py", EVAL_WEATHER),
("function-calling/azure.py", EVAL_WEATHER),
("function-calling/fireworks.py", EVAL_WEATHER),
("function-calling/nvidia.py", EVAL_WEATHER),
("function-calling/cerebras.py", EVAL_WEATHER),
("function-calling/openrouter.py", EVAL_WEATHER),
("function-calling/perplexity.py", EVAL_WEATHER),
("function-calling/google-vertex.py", EVAL_WEATHER),
("function-calling/qwen.py", EVAL_WEATHER),
("function-calling/aws.py", EVAL_WEATHER),
("function-calling/sambanova.py", EVAL_WEATHER),
("function-calling/aws.py", EVAL_WEATHER_AND_RESTAURANT),
("function-calling/nebius.py", EVAL_WEATHER),
("function-calling/mistral.py", EVAL_WEATHER),
("function-calling/sarvam.py", EVAL_WEATHER),
("function-calling/novita.py", EVAL_WEATHER),
# Video
("services/function-calling/anthropic-video.py", EVAL_VISION_CAMERA),
("services/function-calling/aws-video.py", EVAL_VISION_CAMERA),
("services/function-calling/google-video.py", EVAL_VISION_CAMERA),
("services/function-calling/moondream-video.py", EVAL_VISION_CAMERA),
("services/function-calling/openai-video.py", EVAL_VISION_CAMERA),
("services/function-calling/openai-responses-video.py", EVAL_VISION_CAMERA),
("services/function-calling/openai-responses-video-http.py", EVAL_VISION_CAMERA),
("function-calling/anthropic-video.py", EVAL_VISION_CAMERA),
("function-calling/aws-video.py", EVAL_VISION_CAMERA),
("function-calling/google-video.py", EVAL_VISION_CAMERA),
("function-calling/moondream-video.py", EVAL_VISION_CAMERA),
("function-calling/openai-video.py", EVAL_VISION_CAMERA),
("function-calling/openai-responses-video.py", EVAL_VISION_CAMERA),
("function-calling/openai-responses-video-http.py", EVAL_VISION_CAMERA),
# Currently not working.
# ("services/function-calling/together.py", EVAL_WEATHER),
# ("services/function-calling/deepseek.py", EVAL_WEATHER),
# ("services/function-calling/gemini-openai-format.py", EVAL_WEATHER),
# ("function-calling/together.py", EVAL_WEATHER),
# ("function-calling/deepseek.py", EVAL_WEATHER),
# ("function-calling/gemini-openai-format.py", EVAL_WEATHER),
]
TESTS_FEATURES = [
@@ -257,7 +257,7 @@ TESTS_THINKING_AND_MCP = [
]
TESTS = [
*TESTS_SPEECH,
*TESTS_VOICE,
*TESTS_VISION,
*TESTS_FUNCTION_CALLING,
*TESTS_FEATURES,