Merge pull request #340 from pipecat-ai/lewis/silero-vad-via-pip

Install Silero VAD via pip

examples/deployment/flyio-example/Dockerfile

@@ -9,8 +9,5 @@ COPY *.py .
 COPY ./requirements.txt requirements.txt
 RUN pip3 install --no-cache-dir --upgrade -r requirements.txt
 
-# Install models
-RUN python3 install_deps.py
-
 # Start the FastAPI server
 CMD python3 bot_runner.py --port ${FAST_API_PORT}

examples/deployment/flyio-example/README.md

@@ -2,8 +2,6 @@
 
 This project modifies the `bot_runner.py` server to launch a new machine for each user session. This is a recommended approach for production vs. running shell processess as your deployment will quickly run out of system resources under load.
 
-To speed up machine boot times, we also download and cache Silero VAD as part of the Dockerfile (`install_deps.py`). If you are using other custom models, you can add them here too.
-
 For this example, we are using Daily as a WebRTC transport and provisioning a new room and token for each session. You can use another transport, such as WebSockets, by modifying the `bot.py` and `bot_runner.py` files accordingly.
 
 ## Setting up your fly.io deployment
@@ -14,7 +12,7 @@ You can copy the `example-fly.toml` as a reference. Be sure to change the app na
 
 ### Create your .env file
 
-Copy the base `env.example` to `.env` and enter the necessary API keys. 
+Copy the base `env.example` to `.env` and enter the necessary API keys.
 
 `FLY_APP_NAME` should match that in the `fly.toml` file.
 
@@ -32,7 +30,6 @@ Note: you can do this manually via the fly.io dashboard under the "secrets" sub-
 
 `fly deploy`
 
-
 ## Connecting to your bot
 
 Send a post request to your running fly.io instance:
@@ -40,4 +37,3 @@ Send a post request to your running fly.io instance:
 `curl --location --request POST 'https://YOUR_FLY_APP_NAME/start_bot'`
 
 This request will wait until the machine enters into a `starting` state, before returning the a room URL and token to join.
-

examples/deployment/flyio-example/install_deps.py

@@ -1,4 +0,0 @@
-import torch
-
-# Download (cache) the Silero VAD model
-torch.hub.load(repo_or_dir='snakers4/silero-vad', model='silero_vad', force_reload=True)

pyproject.toml

@@ -50,7 +50,7 @@ moondream = [ "einops~=0.8.0", "timm~=0.9.16", "transformers~=4.40.2" ]
 openai = [ "openai~=1.35.0" ]
 openpipe = [ "openpipe~=4.18.0" ]
 playht = [ "pyht~=0.0.28" ]
-silero = [ "torch~=2.3.1", "torchaudio~=2.3.1" ]
+silero = [ "silero-vad~=5.1" ]
 websocket = [ "websockets~=12.0", "fastapi~=0.111.0" ]
 whisper = [ "faster-whisper~=1.0.3" ]
 xtts = [ "resampy~=0.4.3" ]

src/pipecat/vad/silero.py

@@ -15,12 +15,8 @@ from pipecat.vad.vad_analyzer import VADAnalyzer, VADParams, VADState
 from loguru import logger
 
 try:
+    from silero_vad import load_silero_vad
     import torch
-    # We don't use torchaudio here, but we need to try importing it because
-    # Silero uses it.
-    import torchaudio
-
-    torch.set_num_threads(1)
 
 except ModuleNotFoundError as e:
     logger.error(f"Exception: {e}")
@@ -37,10 +33,6 @@ class SileroVADAnalyzer(VADAnalyzer):
             self,
             *,
             sample_rate: int = 16000,
-            version: str = "v5.0",
-            force_reload: bool = False,
-            skip_validation: bool = True,
-            trust_repo: bool = True,
             params: VADParams = VADParams()):
         super().__init__(sample_rate=sample_rate, num_channels=1, params=params)
 
@@ -49,11 +41,7 @@ class SileroVADAnalyzer(VADAnalyzer):
 
         logger.debug("Loading Silero VAD model...")
 
-        (self._model, _) = torch.hub.load(repo_or_dir=f"snakers4/silero-vad:{version}",
-                                          model="silero_vad",
-                                          force_reload=force_reload,
-                                          skip_validation=skip_validation,
-                                          trust_repo=trust_repo)
+        self._model = load_silero_vad()
 
         self._last_reset_time = 0
 
@@ -94,20 +82,12 @@ class SileroVAD(FrameProcessor):
             self,
             *,
             sample_rate: int = 16000,
-            version: str = "v5.0",
-            force_reload: bool = False,
-            skip_validation: bool = True,
-            trust_repo: bool = True,
             vad_params: VADParams = VADParams(),
             audio_passthrough: bool = False):
         super().__init__()
 
         self._vad_analyzer = SileroVADAnalyzer(
             sample_rate=sample_rate,
-            version=version,
-            force_reload=force_reload,
-            skip_validation=skip_validation,
-            trust_repo=trust_repo,
             params=vad_params)
         self._audio_passthrough = audio_passthrough