Integrate Whisper AI into agl-service-voiceagent

V1:
  - Integrated Whisper AI for speech-to-text functionality into
    agl-service-voiceagent.
  - Add support for both online and offline mode.
  - Implemented a gRPC-based connection for online mode between
    Whisper ASR service and voice-agent service.

V2:
  - Update kuksa-interface
  - Add whisper-cpp for speech-to-text functionality
  - Add support to control media using mpd
  - Fix audio recorder

Bug-AGL: SPEC-5200
Change-Id: I2661ae61ba2c3283bcfde26d6e4f498270240b19
Signed-off-by: Anuj Solanki <anuj603362@gmail.com>

1 files changed, 88 insertions, 0 deletions

diff --git a/agl_service_voiceagent/utils/stt_online_service.py b/agl_service_voiceagent/utils/stt_online_service.py
new file mode 100644
index 0000000..7bbdc5d
--- /dev/null
+++ b/agl_service_voiceagent/utils/stt_online_service.py
@@ -0,0 +1,88 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright (c) 2023 Malik Talha
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import grpc
+import sys
+sys.path.append("../")
+from agl_service_voiceagent.generated import audio_processing_pb2
+from agl_service_voiceagent.generated import audio_processing_pb2_grpc
+
+class STTOnlineService:
+    """
+    STTOnlineService class is used to connect to an online gPRC based Whisper ASR service.
+    """
+    def __init__(self, server_address, server_port,server_timeout=5):
+        """
+        Initialize the online speech-to-text service.
+
+        Args:
+            server_ip (str): The IP address of the online speech-to-text service.
+            server_port (int): The port number of the online speech-to-text service.
+            server_timeout (int, optional): The timeout value in seconds (default is 5).
+        """
+        self.server_address = server_address
+        self.server_port = server_port
+        self.server_timeout = server_timeout
+        self.client = None
+        self.initialized = False
+
+
+    def initialize_connection(self):
+        """
+        Initialize the connection to the online speech-to-text service.
+        """
+        try:
+            channel = grpc.insecure_channel(f"{self.server_address}:{self.server_port}")
+            self.client = audio_processing_pb2_grpc.AudioProcessingStub(channel)
+            self.initialized = True
+            print("STTOnlineService initialized with server address:",self.server_address,"and port:",self.server_port,"and timeout:",self.server_timeout,"seconds.")
+        except Exception as e:
+            print("Error initializing online speech-to-text service:",e)
+            self.initialized = False
+        return self.initialized
+
+    def close_connection(self):
+        """
+        Close the connection to the online speech-to-text service.
+        """
+        self.client = None
+        self.initialized = False
+        return not self.initialized
+    
+    def recognize_audio(self, audio_file):
+        """
+        Recognize speech from audio data.
+
+        Args:
+            audio_data (bytes): Audio data to process.
+
+        Returns:
+            str: The recognized text.
+        """
+        if not self.initialized:
+            print("STTOnlineService not initialized.")
+            return None
+        try:
+            with open(audio_file, 'rb') as audio_file:
+                audio_data = audio_file.read()
+            request = audio_processing_pb2.AudioRequest(audio_data=audio_data)
+            response = self.client.ProcessAudio(request,timeout=self.server_timeout)
+            return response.text
+        except Exception as e:
+            print("Error recognizing audio:",e)
+            return None
+
+    
\ No newline at end of file