dv-pipecat-ai 0.0.93.dev87__tar.gz → 0.0.93.dev91__tar.gz

{dv_pipecat_ai-0.0.93.dev87/src/dv_pipecat_ai.egg-info → dv_pipecat_ai-0.0.93.dev91}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dv-pipecat-ai
-Version: 0.0.93.dev87
+Version: 0.0.93.dev91
 Summary: An open source framework for voice (and multimodal) assistants
 License-Expression: BSD-2-Clause
 Project-URL: Source, https://github.com/pipecat-ai/pipecat

{dv_pipecat_ai-0.0.93.dev87 → dv_pipecat_ai-0.0.93.dev91/src/dv_pipecat_ai.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dv-pipecat-ai
-Version: 0.0.93.dev87
+Version: 0.0.93.dev91
 Summary: An open source framework for voice (and multimodal) assistants
 License-Expression: BSD-2-Clause
 Project-URL: Source, https://github.com/pipecat-ai/pipecat

{dv_pipecat_ai-0.0.93.dev87 → dv_pipecat_ai-0.0.93.dev91}/src/dv_pipecat_ai.egg-info/SOURCES.txt

@@ -256,6 +256,7 @@ src/pipecat/serializers/tatatele.py
 src/pipecat/serializers/telnyx.py
 src/pipecat/serializers/twilio.py
 src/pipecat/serializers/vi.py
+src/pipecat/serializers/voicelink.py
 src/pipecat/services/__init__.py
 src/pipecat/services/ai_service.py
 src/pipecat/services/ai_services.py

{dv_pipecat_ai-0.0.93.dev87 → dv_pipecat_ai-0.0.93.dev91}/src/pipecat/serializers/__init__.py

@@ -11,6 +11,7 @@ from .telnyx import TelnyxFrameSerializer
 from .tatatele import TataTeleFrameSerializer
 from .twilio import TwilioFrameSerializer
 from .vi import VIFrameSerializer
+from .voicelink import VoiceLinkFrameSerializer
 
 __all__ = [
     "FrameSerializer",
@@ -27,6 +28,7 @@ __all__ = [
     "TelnyxFrameSerializer",
     "TwilioFrameSerializer",
     "VIFrameSerializer",
+    "VoiceLinkFrameSerializer",
 ]
 
 # Optional imports

dv_pipecat_ai-0.0.93.dev91/src/pipecat/serializers/voicelink.py

@@ -0,0 +1,126 @@
+"""VoiceLink bidirectional audio streaming WebSocket protocol serializer for Pipecat."""
+
+import base64
+import json
+from typing import Optional
+
+from loguru import logger
+from pydantic import BaseModel
+
+from pipecat.audio.dtmf.types import KeypadEntry
+from pipecat.audio.utils import alaw_to_pcm, create_stream_resampler, pcm_to_alaw
+from pipecat.frames.frames import (
+    AudioRawFrame,
+    Frame,
+    InputAudioRawFrame,
+    InputDTMFFrame,
+    InterruptionFrame,
+    OutputTransportMessageFrame,
+    OutputTransportMessageUrgentFrame,
+    StartFrame,
+)
+from pipecat.serializers.base_serializer import FrameSerializer, FrameSerializerType
+
+
+class VoiceLinkFrameSerializer(FrameSerializer):
+    """Serializer for VoiceLink bidirectional audio streaming WebSocket protocol.
+
+    This serializer handles converting between Pipecat frames and VoiceLink's WebSocket
+    media streams protocol. It supports audio conversion and DTMF events.
+
+    Key differences from TataTele/Twilio:
+    - JSON fields use snake_case (stream_sid, call_sid) not camelCase
+    - Audio: G.711 A-law (PCMA), 8000 Hz, 8-bit mono
+    - Media payloads sent TO VoiceLink must be multiples of 160 bytes
+    - Events: connected -> start -> media -> stop
+    """
+
+    class InputParams(BaseModel):
+        voicelink_sample_rate: int = 8000
+        sample_rate: Optional[int] = None
+
+    PAYLOAD_ALIGNMENT = 160
+
+    def __init__(self, stream_sid: str, params: Optional[InputParams] = None):
+        self._params = params or VoiceLinkFrameSerializer.InputParams()
+        self._stream_sid = stream_sid
+        self._voicelink_sample_rate = self._params.voicelink_sample_rate
+        self._sample_rate = 0
+        self._input_resampler = create_stream_resampler()
+        self._output_resampler = create_stream_resampler()
+        self._output_buffer = bytearray()
+
+    @property
+    def type(self) -> FrameSerializerType:
+        return FrameSerializerType.TEXT
+
+    async def setup(self, frame: StartFrame):
+        self._sample_rate = self._params.sample_rate or frame.audio_in_sample_rate
+
+    async def serialize(self, frame: Frame) -> str | bytes | None:
+        if isinstance(frame, InterruptionFrame):
+            # Drop any unsent sub-160-byte remainder so pre-interruption audio isn't prepended to
+            # the next AudioRawFrame after barge-in (the `clear` event only flushes VoiceLink's
+            # queued audio, not this local alignment buffer).
+            self._output_buffer = bytearray()
+            answer = {"event": "clear", "stream_sid": self._stream_sid}
+            return json.dumps(answer)
+        elif isinstance(frame, AudioRawFrame):
+            data = frame.audio
+
+            serialized_data = await pcm_to_alaw(
+                data, frame.sample_rate, self._voicelink_sample_rate, self._output_resampler
+            )
+            if serialized_data is None or len(serialized_data) == 0:
+                return None
+
+            self._output_buffer.extend(serialized_data)
+            aligned_length = (
+                len(self._output_buffer) // self.PAYLOAD_ALIGNMENT * self.PAYLOAD_ALIGNMENT
+            )
+            if aligned_length == 0:
+                return None
+
+            to_send = bytes(self._output_buffer[:aligned_length])
+            self._output_buffer = self._output_buffer[aligned_length:]
+
+            payload = base64.b64encode(to_send).decode("utf-8")
+            answer = {
+                "event": "media",
+                "stream_sid": self._stream_sid,
+                "media": {"payload": payload},
+            }
+
+            return json.dumps(answer)
+        elif isinstance(frame, (OutputTransportMessageFrame, OutputTransportMessageUrgentFrame)):
+            return json.dumps(frame.message)
+
+        return None
+
+    async def deserialize(self, data: str | bytes) -> Frame | None:
+        message = json.loads(data)
+
+        if message.get("event") == "media":
+            payload_base64 = message["media"]["payload"]
+            payload = base64.b64decode(payload_base64)
+
+            deserialized_data = await alaw_to_pcm(
+                payload, self._voicelink_sample_rate, self._sample_rate, self._input_resampler
+            )
+            if deserialized_data is None or len(deserialized_data) == 0:
+                return None
+
+            audio_frame = InputAudioRawFrame(
+                audio=deserialized_data, num_channels=1, sample_rate=self._sample_rate
+            )
+            return audio_frame
+        elif message.get("event") == "dtmf":
+            digit = message.get("dtmf", {}).get("digit")
+            if digit:
+                try:
+                    return InputDTMFFrame(KeypadEntry(digit))
+                except ValueError:
+                    logger.warning(f"Invalid DTMF digit received: {digit}")
+                    return None
+        else:
+            return None