dv-pipecat-ai 0.0.85.dev818__py3-none-any.whl → 0.0.85.dev858__py3-none-any.whl

pipecat/serializers/vi.py

@@ -0,0 +1,326 @@
+#
+# Copyright (c) 2024–2025, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Vodafone Idea (VI) WebSocket frame serializer for audio streaming and call management."""
+
+import base64
+import json
+from datetime import datetime, timezone
+from typing import Optional
+
+from loguru import logger
+from pydantic import BaseModel
+
+from pipecat.audio.utils import create_default_resampler
+from pipecat.frames.frames import (
+    AudioRawFrame,
+    CancelFrame,
+    EndFrame,
+    Frame,
+    InputAudioRawFrame,
+    InputDTMFFrame,
+    KeypadEntry,
+    StartFrame,
+    StartInterruptionFrame,
+    TransportMessageFrame,
+    TransportMessageUrgentFrame,
+)
+from pipecat.serializers.base_serializer import FrameSerializer, FrameSerializerType
+
+
+class VIFrameSerializer(FrameSerializer):
+    """Serializer for Vodafone Idea (VI) WebSocket protocol.
+
+    This serializer handles converting between Pipecat frames and VI's WebSocket
+    protocol for bidirectional audio streaming. It supports audio conversion, DTMF events,
+    and real-time communication with VI telephony systems.
+
+    VI WebSocket protocol requirements:
+    - PCM audio format at 8kHz sample rate
+    - 16-bit Linear PCM encoding
+    - Base64 encoded audio payloads
+    - JSON message format for control and media events
+    - Bitrate: 128 Kbps
+
+    Events (VI → Endpoint):
+    - connected: WebSocket connection established
+    - start: Stream session started with call/stream IDs
+    - media: Audio data in Base64-encoded PCM
+    - dtmf: Keypad digit pressed
+    - stop: Stream ended
+    - mark: Audio playback checkpoint confirmation
+
+    Events (Endpoint → VI):
+    - media: Send audio back to VI
+    - mark: Request acknowledgment for audio playback
+    - clear: Clear queued audio (interruption)
+    - exit: Terminate session gracefully
+    """
+
+    class InputParams(BaseModel):
+        """Configuration parameters for VIFrameSerializer.
+
+        Attributes:
+            vi_sample_rate: Sample rate used by VI, defaults to 8000 Hz (telephony standard).
+            sample_rate: Optional override for pipeline input sample rate.
+            auto_hang_up: Whether to automatically terminate call on EndFrame.
+        """
+
+        vi_sample_rate: int = 8000
+        sample_rate: Optional[int] = None
+        auto_hang_up: bool = False
+
+    def __init__(
+        self,
+        stream_id: str,
+        call_id: Optional[str] = None,
+        params: Optional[InputParams] = None,
+    ):
+        """Initialize the VIFrameSerializer.
+
+        Args:
+            stream_id: The VI stream identifier.
+            call_id: The associated VI call identifier.
+            params: Configuration parameters.
+        """
+        self._stream_id = stream_id
+        self._call_id = call_id
+        self._params = params or VIFrameSerializer.InputParams()
+
+        self._vi_sample_rate = self._params.vi_sample_rate
+        self._sample_rate = 0  # Pipeline input rate
+        self._call_ended = False
+
+        self._resampler = create_default_resampler()
+
+    @property
+    def type(self) -> FrameSerializerType:
+        """Gets the serializer type.
+
+        Returns:
+            The serializer type as TEXT for JSON WebSocket messages.
+        """
+        return FrameSerializerType.TEXT
+
+    async def setup(self, frame: StartFrame):
+        """Sets up the serializer with pipeline configuration.
+
+        Args:
+            frame: The StartFrame containing pipeline configuration.
+        """
+        self._sample_rate = self._params.sample_rate or frame.audio_in_sample_rate
+
+    async def serialize(self, frame: Frame) -> str | bytes | None:
+        """Serializes a Pipecat frame to VI WebSocket format.
+
+        Handles conversion of various frame types to VI WebSocket messages.
+        For EndFrames, initiates call termination if auto_hang_up is enabled.
+
+        Args:
+            frame: The Pipecat frame to serialize.
+
+        Returns:
+            Serialized data as JSON string, or None if the frame isn't handled.
+        """
+        if (
+            self._params.auto_hang_up
+            and not self._call_ended
+            and isinstance(frame, (EndFrame, CancelFrame))
+        ):
+            self._call_ended = True
+            # Return the exit event to terminate the VI session
+            return await self._send_exit_event()
+
+        elif isinstance(frame, StartInterruptionFrame):
+            # Clear/interrupt command for VI - clears queued audio
+            message = {
+                "event": "clear",
+                "stream_id": self._stream_id,
+                "call_id": self._call_id,
+            }
+            logger.debug(f"VI: Sending clear event for stream_id: {self._stream_id}")
+            return json.dumps(message)
+
+        elif isinstance(frame, AudioRawFrame):
+            if self._call_ended:
+                logger.debug("VI SERIALIZE: Skipping audio - call has ended")
+                return None
+
+            # Convert PCM audio to VI format
+            data = frame.audio
+
+            # Resample to VI sample rate (8kHz)
+            serialized_data = await self._resampler.resample(
+                data, frame.sample_rate, self._vi_sample_rate
+            )
+
+            # Encode as base64 for transmission
+            payload = base64.b64encode(serialized_data).decode("ascii")
+
+            # VI expects media event format with Base64-encoded PCM audio
+            timestamp = datetime.now(timezone.utc).isoformat().replace("+00:00", "Z")
+
+            message = {
+                "event": "media",
+                "stream_id": self._stream_id,
+                "media": {
+                    "timestamp": timestamp,
+                    "chunk": len(serialized_data),  # Chunk size in bytes
+                    "payload": payload,
+                },
+            }
+
+            logger.debug(f"VI: Sending media event {message} for stream_id: {self._stream_id}")
+
+            return json.dumps(message)
+
+        elif isinstance(frame, (TransportMessageFrame, TransportMessageUrgentFrame)):
+            # Pass through transport messages (for mark events, etc.)
+            return json.dumps(frame.message)
+
+        return None
+
+    async def _send_exit_event(self):
+        """Send an exit event to VI to terminate the session gracefully.
+
+        This method is called when auto_hang_up is enabled and an EndFrame or
+        CancelFrame is received. The exit event allows IVR logic to continue
+        after the WebSocket session ends.
+        """
+        try:
+            exit_event = {
+                "event": "exit",
+                "stream_id": self._stream_id,
+                "call_id": self._call_id,
+                "timestamp": datetime.now(timezone.utc).isoformat().replace("+00:00", "Z"),
+            }
+
+            logger.info(
+                f"VI auto_hang_up: Sending exit event for stream_id: {self._stream_id}, call_id: {self._call_id}"
+            )
+            return json.dumps(exit_event)
+        except Exception as e:
+            logger.error(f"VI auto_hang_up: Failed to create exit event: {e}")
+            return None
+
+    async def deserialize(self, data: str | bytes) -> Frame | None:
+        """Deserializes VI WebSocket data to Pipecat frames.
+
+        Handles conversion of VI media events to appropriate Pipecat frames.
+
+        Args:
+            data: The raw WebSocket data from VI.
+
+        Returns:
+            A Pipecat frame corresponding to the VI event, or None if unhandled.
+        """
+        try:
+            message = json.loads(data)
+        except json.JSONDecodeError:
+            logger.error(f"Invalid JSON received from VI: {data}")
+            return None
+
+        # Log all incoming events for debugging and monitoring
+        event = message.get("event")
+        logger.debug(
+            f"VI INCOMING EVENT: {event} - stream_id: {self._stream_id}, call_id: {self._call_id}"
+        )
+
+        if event == "media":
+            # Handle incoming audio data from VI
+            media = message.get("media", {})
+            payload_base64 = media.get("payload")
+
+            if not payload_base64:
+                logger.warning("VI DESERIALIZE: No payload in VI media message")
+                return None
+
+            try:
+                payload = base64.b64decode(payload_base64)
+                chunk_size = len(payload)
+
+                # Log chunk info (optional)
+                logger.debug(
+                    f"VI DESERIALIZE: Received audio from VI - {chunk_size} bytes at {self._vi_sample_rate}Hz"
+                )
+
+            except Exception as e:
+                logger.error(f"VI DESERIALIZE: Error decoding VI audio payload: {e}")
+                return None
+
+            # Convert from VI sample rate (8kHz) to pipeline sample rate
+            deserialized_data = await self._resampler.resample(
+                payload,
+                self._vi_sample_rate,
+                self._sample_rate,
+            )
+
+            audio_frame = InputAudioRawFrame(
+                audio=deserialized_data,
+                num_channels=1,  # VI uses mono audio
+                sample_rate=self._sample_rate,
+            )
+            return audio_frame
+
+        elif event == "dtmf":
+            # Handle DTMF events
+            dtmf_data = message.get("dtmf", {})
+            digit = dtmf_data.get("digit")
+
+            if digit:
+                try:
+                    logger.info(f"VI: Received DTMF digit: {digit}")
+                    return InputDTMFFrame(KeypadEntry(digit))
+                except ValueError:
+                    logger.warning(f"Invalid DTMF digit from VI: {digit}")
+                    return None
+
+        elif event == "connected":
+            # Handle connection event
+            logger.info(f"VI connection established: {message}")
+            return None
+
+        elif event == "start":
+            # Handle stream start event
+            logger.info(f"VI stream started: {message}")
+            return None
+
+        elif event == "stop":
+            # Handle stream stop event
+            logger.info(f"VI stream stopped: {message}")
+            # Don't end the call here, wait for explicit exit or call end
+            return None
+
+        elif event == "mark":
+            # Handle mark event - checkpoint confirming audio playback completion
+            mark_data = message.get("mark", {})
+            mark_name = mark_data.get("name", "unknown")
+            logger.info(f"VI mark event received: {mark_name}")
+            # Mark events are informational, no frame to return
+            return None
+
+        elif event == "error":
+            # Handle error events
+            error_msg = message.get("error", "Unknown error")
+            logger.error(f"VI error: {error_msg}")
+            return None
+
+        elif event == "exit":
+            # Handle exit event from VI
+            logger.info("VI exit event received - terminating session")
+            self._call_ended = True
+            return CancelFrame()
+
+        elif event == "call_end" or event == "callEnd":
+            # Handle call end event (if VI sends this)
+            logger.info("VI call end event received")
+            self._call_ended = True
+            return CancelFrame()
+
+        else:
+            logger.debug(f"VI UNHANDLED EVENT: {event}")
+
+        return None

pipecat/services/cartesia/tts.py

@@ -15,7 +15,6 @@ from typing import AsyncGenerator, List, Literal, Optional, Union
 from loguru import logger
 from pydantic import BaseModel, Field
 
-
 from pipecat.frames.frames import (
     CancelFrame,
     EndFrame,
@@ -49,6 +48,26 @@ except ModuleNotFoundError as e:
     raise Exception(f"Missing module: {e}")
 
 
+class GenerationConfig(BaseModel):
+    """Configuration for Cartesia Sonic-3 generation parameters.
+
+    Sonic-3 interprets these parameters as guidance to ensure natural speech.
+    Test against your content for best results.
+
+    Parameters:
+        volume: Volume multiplier for generated speech. Valid range: [0.5, 2.0]. Default is 1.0.
+        speed: Speed multiplier for generated speech. Valid range: [0.6, 1.5]. Default is 1.0.
+        emotion: Single emotion string to guide the emotional tone. Examples include neutral,
+            angry, excited, content, sad, scared. Over 60 emotions are supported. For best
+            results, use with recommended voices: Leo, Jace, Kyle, Gavin, Maya, Tessa, Dana,
+            and Marian.
+    """
+
+    volume: Optional[float] = None
+    speed: Optional[float] = None
+    emotion: Optional[str] = None
+
+
 def language_to_cartesia_language(language: Language) -> Optional[str]:
     """Convert a Language enum to Cartesia language code.
 
@@ -74,6 +93,33 @@ def language_to_cartesia_language(language: Language) -> Optional[str]:
         Language.SV: "sv",
         Language.TR: "tr",
         Language.ZH: "zh",
+        Language.TL: "tl",
+        Language.BG: "bg",
+        Language.RO: "ro",
+        Language.AR: "ar",
+        Language.CS: "cs",
+        Language.EL: "el",
+        Language.FI: "fi",
+        Language.HR: "hr",
+        Language.MS: "ms",
+        Language.SK: "sk",
+        Language.DA: "da",
+        Language.TA: "ta",
+        Language.UK: "uk",
+        Language.HU: "hu",
+        Language.NO: "no",
+        Language.VI: "vi",
+        Language.BN: "bn",
+        Language.TH: "th",
+        Language.HE: "he",
+        Language.KA: "ka",
+        Language.ID: "id",
+        Language.TE: "te",
+        Language.GU: "gu",
+        Language.KN: "kn",
+        Language.ML: "ml",
+        Language.MR: "mr",
+        Language.PA: "pa",
     }
 
     result = BASE_LANGUAGES.get(language)
@@ -102,16 +148,20 @@ class CartesiaTTSService(AudioContextWordTTSService):
 
         Parameters:
             language: Language to use for synthesis.
-            speed: Voice speed control.
-            emotion: List of emotion controls.
+            speed: Voice speed control for non-Sonic-3 models (literal values).
+            emotion: List of emotion controls for non-Sonic-3 models.
 
                 .. deprecated:: 0.0.68
                         The `emotion` parameter is deprecated and will be removed in a future version.
+
+            generation_config: Generation configuration for Sonic-3 models. Includes volume,
+                speed (numeric), and emotion (string) parameters.
         """
 
         language: Optional[Language] = Language.EN
         speed: Optional[Literal["slow", "normal", "fast"]] = None
         emotion: Optional[List[str]] = []
+        generation_config: Optional[GenerationConfig] = None
 
     def __init__(
         self,
@@ -120,7 +170,7 @@ class CartesiaTTSService(AudioContextWordTTSService):
         voice_id: str,
         cartesia_version: str = "2025-04-16",
         url: str = "wss://api.cartesia.ai/tts/websocket",
-        model: str = "sonic-2",
+        model: str = "sonic-3",
         sample_rate: Optional[int] = None,
         encoding: str = "pcm_s16le",
         container: str = "raw",
@@ -136,7 +186,7 @@ class CartesiaTTSService(AudioContextWordTTSService):
             voice_id: ID of the voice to use for synthesis.
             cartesia_version: API version string for Cartesia service.
             url: WebSocket URL for Cartesia TTS API.
-            model: TTS model to use (e.g., "sonic-2").
+            model: TTS model to use (e.g., "sonic-3").
             sample_rate: Audio sample rate. If None, uses default.
             encoding: Audio encoding format.
             container: Audio container format.
@@ -180,6 +230,7 @@ class CartesiaTTSService(AudioContextWordTTSService):
             else "en",
             "speed": params.speed,
             "emotion": params.emotion,
+            "generation_config": params.generation_config,
         }
         self.set_model_name(model)
         self.set_voice(voice_id)
@@ -298,6 +349,11 @@ class CartesiaTTSService(AudioContextWordTTSService):
         if self._settings["speed"]:
             msg["speed"] = self._settings["speed"]
 
+        if self._settings["generation_config"]:
+            msg["generation_config"] = self._settings["generation_config"].model_dump(
+                exclude_none=True
+            )
+
         return json.dumps(msg)
 
     async def start(self, frame: StartFrame):
@@ -419,7 +475,6 @@ class CartesiaTTSService(AudioContextWordTTSService):
                 logger.error(f"{self} error: {msg}")
                 await self.push_frame(TTSStoppedFrame())
                 await self.stop_all_metrics()
-
                 await self.push_error(ErrorFrame(f"{self} error: {msg['error']}"))
                 self._context_id = None
             else:
@@ -484,23 +539,27 @@ class CartesiaHttpTTSService(TTSService):
 
         Parameters:
             language: Language to use for synthesis.
-            speed: Voice speed control.
-            emotion: List of emotion controls.
+            speed: Voice speed control for non-Sonic-3 models (literal values).
+            emotion: List of emotion controls for non-Sonic-3 models.
 
                 .. deprecated:: 0.0.68
                         The `emotion` parameter is deprecated and will be removed in a future version.
+
+            generation_config: Generation configuration for Sonic-3 models. Includes volume,
+                speed (numeric), and emotion (string) parameters.
         """
 
         language: Optional[Language] = Language.EN
         speed: Optional[Literal["slow", "normal", "fast"]] = None
         emotion: Optional[List[str]] = Field(default_factory=list)
+        generation_config: Optional[GenerationConfig] = None
 
     def __init__(
         self,
         *,
         api_key: str,
         voice_id: str,
-        model: str = "sonic-2",
+        model: str = "sonic-3",
         base_url: str = "https://api.cartesia.ai",
         cartesia_version: str = "2024-11-13",
         sample_rate: Optional[int] = None,
@@ -514,7 +573,7 @@ class CartesiaHttpTTSService(TTSService):
         Args:
             api_key: Cartesia API key for authentication.
             voice_id: ID of the voice to use for synthesis.
-            model: TTS model to use (e.g., "sonic-2").
+            model: TTS model to use (e.g., "sonic-3").
             base_url: Base URL for Cartesia HTTP API.
             cartesia_version: API version string for Cartesia service.
             sample_rate: Audio sample rate. If None, uses default.
@@ -541,6 +600,7 @@ class CartesiaHttpTTSService(TTSService):
             else "en",
             "speed": params.speed,
             "emotion": params.emotion,
+            "generation_config": params.generation_config,
         }
         self.set_voice(voice_id)
         self.set_model_name(model)
@@ -634,6 +694,11 @@ class CartesiaHttpTTSService(TTSService):
             if self._settings["speed"]:
                 payload["speed"] = self._settings["speed"]
 
+            if self._settings["generation_config"]:
+                payload["generation_config"] = self._settings["generation_config"].model_dump(
+                    exclude_none=True
+                )
+
             yield TTSStartedFrame()
 
             session = await self._client._get_session()