minor cleaning

backend/routers/transcribe_v2.py

@@ -59,10 +59,10 @@ def get_model_name(value):
 
 async def _websocket_util(
         websocket: WebSocket, uid: str, language: str = 'en', sample_rate: int = 8000, codec: str = 'pcm8',
-        channels: int = 1, include_speech_profile: bool = True, new_memory_watch: bool = False,
+        include_speech_profile: bool = True,
         # stt_service: STTService = STTService.deepgram,
 ):
-    print('websocket_endpoint', uid, language, sample_rate, codec, channels, include_speech_profile, new_memory_watch)
+    print('websocket_endpoint', uid, language, sample_rate, codec, include_speech_profile)
 
     # Not when comes from the phone, and only Friend's with 1.0.4
     if language == 'en' and sample_rate == 16000 and codec == 'opus':
@@ -181,11 +181,11 @@ def stream_transcript(segments, _):
         # DEEPGRAM
         if stt_service == STTService.deepgram:
             deepgram_socket = await process_audio_dg(
-                stream_transcript, 1, language, sample_rate, channels, preseconds=speech_profile_duration
+                stream_transcript, 1, language, sample_rate, 1, preseconds=speech_profile_duration
             )
             if speech_profile_duration:
                 deepgram_socket2 = await process_audio_dg(
-                    stream_transcript, 2, language, sample_rate, channels
+                    stream_transcript, 2, language, sample_rate, 1
                 )
 
                 async def deepgram_socket_send(data):
@@ -213,7 +213,7 @@ async def deepgram_socket_send(data):
         await websocket.close(code=websocket_close_code)
         return
 
-    decoder = opuslib.Decoder(sample_rate, channels)
+    decoder = opuslib.Decoder(sample_rate, 1)
     websocket_active = True
     websocket_close_code = 1001  # Going Away, don't close with good from backend
 
@@ -362,9 +362,9 @@ async def _create_memory():
 @router.websocket("/v2/listen")
 async def websocket_endpoint(
         websocket: WebSocket, uid: str, language: str = 'en', sample_rate: int = 8000, codec: str = 'pcm8',
-        channels: int = 1, include_speech_profile: bool = True, new_memory_watch: bool = False,
+        channels: int = 1, include_speech_profile: bool = True,
         # stt_service: STTService = STTService.deepgram
 ):
     await _websocket_util(
-        websocket, uid, language, sample_rate, codec, channels, include_speech_profile, new_memory_watch,  # stt_service
+        websocket, uid, language, sample_rate, codec, channels, include_speech_profile,  # stt_service
     )

backend/utils/stt/streaming.py

@@ -4,7 +4,7 @@
 from typing import List
 
 import websockets
-from deepgram import DeepgramClient, DeepgramClientOptions, LiveTranscriptionEvents, ListenWebSocketClient
+from deepgram import DeepgramClient, DeepgramClientOptions, LiveTranscriptionEvents
 from deepgram.clients.live.v1 import LiveOptions
 
 import database.notifications as notification_db
@@ -94,8 +94,7 @@ async def send_initial_file(data: List[List[int]], transcript_socket):
 
 
 async def process_audio_dg(
-        stream_transcript, stream_id: int, language: str, sample_rate: int, channels: int,
-        preseconds: int = 0,
+        stream_transcript, stream_id: int, language: str, sample_rate: int, channels: int, preseconds: int = 0,
 ):
     print('process_audio_dg', language, sample_rate, channels, preseconds)
 
@@ -342,7 +341,8 @@ async def on_message():
         raise  # Re-raise the exception to be handled by the caller
 
 
-async def process_audio_speechmatics(stream_transcript, stream_id: int, sample_rate: int, language: str, preseconds: int = 0):
+async def process_audio_speechmatics(stream_transcript, stream_id: int, sample_rate: int, language: str,
+                                     preseconds: int = 0):
     api_key = os.getenv('SPEECHMATICS_API_KEY')
     uri = 'wss://eu2.rt.speechmatics.com/v2'