Added audio streaming parser (#35827)

Author: v-durgeshs

sdk/communication/azure-communication-callautomation/azure/communication/callautomation/_models.py

@@ -4,8 +4,10 @@
 # Licensed under the MIT License. See License.txt in the project root for
 # license information.
 # --------------------------------------------------------------------------
+
 from typing import List, Optional, Union, TYPE_CHECKING
 from enum import Enum
+from datetime import datetime
 from typing_extensions import Literal
 from ._generated.models import (
     CallLocator,
@@ -1062,3 +1064,68 @@ def __init__(
         self.words = words
         self.participant = participant
         self.result_state = result_state
+
+class AudioMetadata:
+    """
+    Metadata for Audio Streaming.
+    :keyword  subscription_id: Audio subscription id.
+    :paramtype  subscription_id: str
+    :keyword encoding: Audio encoding.
+    :paramtype encoding: str
+    :keyword sample_rate: Audio sample rate.
+    :paramtype sample_rate: int
+    :keyword channels: Audio channels.
+    :paramtype channels: int
+    :keyword length: Audio length.
+    :paramtype length: int
+    """
+    subscription_id: str
+    """ Audio subscription id. """
+    encoding: str
+    """ Audio encoding. """
+    sample_rate: int
+    """ Audio sample rate. """
+    channels: int
+    """ Audio channels. """
+    length: int
+    """ Audio length. """
+
+    def __init__(
+            self,
+            *,
+            subscription_id: str,
+            encoding: str,
+            sample_rate: int,
+            channels: int,
+            length: int):
+        self.subscription_id = subscription_id
+        self.encoding = encoding
+        self.sample_rate = sample_rate
+        self.channels = channels
+        self.length = length
+
+class AudioData:
+    """
+    Data for Audio Streaming.
+    :keyword  data: Audio streaming data.
+    :paramtype  data: str
+    :keyword time_stamp: time stamp.
+    :paramtype time_stamp: datetime
+    :keyword is_silent: Is silent.
+    :paramtype is_silent: bool
+    """
+    data: str
+    """ Audio streaming data. """
+    time_stamp : datetime
+    """ Time stamp. """
+    is_silent : bool
+    """ Is silent. """
+    def __init__(
+           self,
+           *,
+           data: str,
+           time_stamp: str,
+           is_silent: bool):
+        self.data = data
+        self.time_stamp = time_stamp
+        self.is_silent = is_silent

sdk/communication/azure-communication-callautomation/azure/communication/callautomation/_streaming_data_parser.py

@@ -7,17 +7,26 @@
 from typing import Union
 import json
 from azure.communication.callautomation._shared.models import identifier_from_raw_id
-from azure.communication.callautomation._models import (TranscriptionMetadata,TranscriptionData,WordData)
+from azure.communication.callautomation._models import (
+    TranscriptionMetadata,
+    TranscriptionData,
+    WordData,
+    AudioMetadata,
+    AudioData)
 
 class StreamingDataParser:
     @staticmethod
-    def parse(packet_data: Union[str, bytes]) -> Union[TranscriptionMetadata, TranscriptionData]:
+    def parse(packet_data: Union[str, bytes]) -> Union[
+        TranscriptionMetadata,
+        TranscriptionData,
+        AudioMetadata,
+        AudioData]:
         """
         Parse the incoming packets.
         :param packet_data: Transcription packet data.
         :type packet_data: Union[str, bytes]
-        :return: Union[TranscriptionMetadata, TranscriptionData]
-        :rtype: TranscriptionMetadata, TranscriptionData
+        :return: Union[TranscriptionMetadata, TranscriptionData, AudioMetadata, AudioData]
+        :rtype: TranscriptionMetadata, TranscriptionData, AudioMetadata, AudioData
         :raises: ValueError
         """
         if isinstance(packet_data, str):
@@ -52,4 +61,20 @@ def parse(packet_data: Union[str, bytes]) -> Union[TranscriptionMetadata, Transc
                 result_state=json_object['transcriptionData']['resultStatus']
             )
             return transcription_data
+        if kind == 'AudioMetadata':
+            audio_metadata = AudioMetadata(
+                subscription_id=json_object['audioMetadata']['subscriptionId'],
+                encoding=json_object['audioMetadata']['encoding'],
+                sample_rate=json_object['audioMetadata']['sampleRate'],
+                channels=json_object['audioMetadata']['channels'],
+                length=json_object['audioMetadata']['length']
+            )
+            return audio_metadata
+        if kind == 'AudioData':
+            audio_data = AudioData(
+                data=json_object['audioData']['data'],
+                time_stamp=json_object['audioData']['timestamp'],
+                is_silent=json_object['audioData']['silent']
+            )
+            return audio_data
         raise ValueError(string_json)

sdk/communication/azure-communication-callautomation/tests/test_streaming_data_parser.py

@@ -6,14 +6,19 @@
 
 import json
 import unittest
-from azure.communication.callautomation._models import (TranscriptionMetadata,TranscriptionData,WordData,TextFormat)
+from azure.communication.callautomation._models import (
+    TranscriptionMetadata,
+    TranscriptionData,
+    AudioMetadata,
+    AudioData)
 from azure.communication.callautomation._streaming_data_parser import StreamingDataParser
 
 class TestStreamDataParser(unittest.TestCase):
     def setUp(self):
         self.transcriptionMetaDataJson = '{"kind":"TranscriptionMetadata","transcriptionMetadata":{"subscriptionId":"0000a000-9999-5555-ae00-cd00e0bc0000","locale":"en-US","callConnectionId":"6d09449c-6677-4f91-8cb7-012c338e6ec1","correlationId":"6d09449c-6677-4f91-8cb7-012c338e6ec1"}}'
         self.transcriptionDataJson = '{"kind":"TranscriptionData","transcriptionData":{"text":"Is everything fine.","format":"display","confidence":0.8138430714607239,"offset":868464674,"duration":11600000,"words":[{"text":"is","offset":868464674,"duration":2400000},{"text":"everything","offset":870864674,"duration":5200000},{"text":"fine","offset":876064674,"duration":4000000}],"participantRawID":"4:+910000000000","resultStatus":"Final"}}'
-
+        self.audioMetadataJson = '{"kind":"AudioMetadata","audioMetadata":{"subscriptionId":"4af370df-3868-461f-8242-91f077a6f8a6","encoding":"PCM","sampleRate":16000,"channels":1,"length":640}}'
+        self.audioDataJson = '{"kind":"AudioData","audioData":{"timestamp":"2024-05-30T06:25:02.948Z","data":"test","silent":false}}'
     def test_parse_binary_to_transcription_metadata(self):
         transcriptionMetaDataBinary = self.transcriptionMetaDataJson.encode('utf-8')
         parsedData = StreamingDataParser.parse(transcriptionMetaDataBinary)
@@ -60,3 +65,37 @@ def validate_transcription_data(self, transcriptionData):
         self.assertEqual(transcriptionData.words[2].offset, 876064674)
         self.assertEqual(transcriptionData.words[2].duration, 4000000)
         self.assertEqual(transcriptionData.participant.raw_id, "4:+910000000000")
+
+    def test_parse_json_to_audio_metadata(self):
+        parsedData = StreamingDataParser.parse(self.audioMetadataJson)
+        self.assertTrue(isinstance(parsedData, AudioMetadata))
+        self.validate_audio_metadata(parsedData)
+
+    def test_parse_binary_to_audio_metadata(self):
+        audioMetadataBinary = self.audioMetadataJson.encode('utf-8')
+        parsedData = StreamingDataParser.parse(audioMetadataBinary)
+        self.assertTrue(isinstance(parsedData, AudioMetadata))
+        self.validate_audio_metadata(parsedData)
+
+    def test_parse_json_to_audio_data(self):
+        parsedData = StreamingDataParser.parse(self.audioDataJson)
+        self.assertTrue(isinstance(parsedData, AudioData))
+        self.validate_audio_data(parsedData)
+
+    def test_parse_binary_to_audio_data(self):
+        audioDataBinary = self.audioDataJson.encode('utf-8')
+        parsedData = StreamingDataParser.parse(audioDataBinary)
+        self.assertTrue(isinstance(parsedData, AudioData))
+        self.validate_audio_data(parsedData)
+        
+    def validate_audio_metadata(self, audioMetadata):
+        self.assertEqual(audioMetadata.subscription_id,'4af370df-3868-461f-8242-91f077a6f8a6')
+        self.assertEqual(audioMetadata.encoding,'PCM')
+        self.assertEqual(audioMetadata.sample_rate,16000)
+        self.assertEqual(audioMetadata.channels,1)
+        self.assertEqual(audioMetadata.length,640)
+    
+    def validate_audio_data(self, audioData):
+        self.assertEqual(audioData.data,"test")
+        self.assertEqual(audioData.is_silent,False)
+