videosdk-plugins-navana 0.0.30__tar.gz → 0.0.32__tar.gz

{videosdk_plugins_navana-0.0.30 → videosdk_plugins_navana-0.0.32}/.gitignore

@@ -2,13 +2,12 @@ myenv/
 venv/
 env/
 __pycache__/
-
+.venv/
 .env
 .env.local
 test_env/
 dist/
 .DS_Store
-
 node_modules/
 credentials.json
 .Python
@@ -16,3 +15,5 @@ build/
 eggs/
 sdist/
 wheels/
+docs/
+agent-sdk-reference/

{videosdk_plugins_navana-0.0.30 → videosdk_plugins_navana-0.0.32}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: videosdk-plugins-navana
-Version: 0.0.30
+Version: 0.0.32
 Summary: VideoSDK Agent Framework plugin for Navana STT services
 Author: videosdk
 License-Expression: Apache-2.0
@@ -16,7 +16,7 @@ Requires-Dist: aiohttp
 Requires-Dist: bodhi-sdk
 Requires-Dist: numpy
 Requires-Dist: scipy>=1.11.0
-Requires-Dist: videosdk-agents>=0.0.30
+Requires-Dist: videosdk-agents>=0.0.32
 Description-Content-Type: text/markdown
 
 # VideoSDK Navana Plugin

{videosdk_plugins_navana-0.0.30 → videosdk_plugins_navana-0.0.32}/pyproject.toml

@@ -32,7 +32,7 @@ classifiers = [
     "Topic :: Scientific/Engineering :: Artificial Intelligence",
 ]
 dependencies = [
-    "videosdk-agents>=0.0.30",
+    "videosdk-agents>=0.0.32",
     "aiohttp",
     "numpy",
     "bodhi-sdk",

{videosdk_plugins_navana-0.0.30 → videosdk_plugins_navana-0.0.32}/videosdk/plugins/navana/stt.py

@@ -13,53 +13,67 @@ try:
 except ImportError:
     SCIPY_AVAILABLE = False
 
+
 class NavanaSTT(BaseSTT):
     """
     VideoSDK Agent Framework STT plugin for Navana's Bodhi API.
-    
+
     This plugin uses the official 'bodhi-sdk' and implements best practices for audio handling,
     including robust stereo-to-mono conversion and event model adaptation.
     """
-    
+
     def __init__(
         self,
         *,
-        customer_id: str | None = None,
         api_key: str | None = None,
+        customer_id: str | None = None,
         model: str = "en-general-v2-8khz",
         language: str = "en",
         input_sample_rate: int = 48000,
     ) -> None:
+        """Initialize the Navana STT plugin.
+
+        Args:
+            api_key (Optional[str], optional): Navana API key. Defaults to None.
+            customer_id (Optional[str], optional): Navana customer ID. Defaults to None.
+            model (str): The model to use for the STT plugin. Defaults to "en-general-v2-8khz".
+            language (str): The language to use for the STT plugin. Defaults to "en".
+            input_sample_rate (int): The input sample rate to use for the STT plugin. Defaults to 48000.
+        """
         super().__init__()
 
         if not SCIPY_AVAILABLE:
-            raise ImportError("The 'scipy' library is not installed. Please install it with 'pip install scipy' to use the NavanaSTT plugin for audio resampling.")
+            raise ImportError(
+                "The 'scipy' library is not installed. Please install it with 'pip install scipy' to use the NavanaSTT plugin for audio resampling.")
 
         self.customer_id = customer_id or os.getenv("NAVANA_CUSTOMER_ID")
         self.api_key = api_key or os.getenv("NAVANA_API_KEY")
-        
+
         if not self.api_key or not self.customer_id:
             raise ValueError(
                 "Navana API key and Customer ID must be provided either through parameters or "
                 "NAVANA_API_KEY/NAVANA_CUSTOMER_ID environment variables."
             )
-        
+
         self.model = model
         self.language = language
         self.input_sample_rate = input_sample_rate
         self.target_sample_rate = 8000
-        
-        self.client = BodhiClient(api_key=self.api_key, customer_id=self.customer_id)
+
+        self.client = BodhiClient(
+            api_key=self.api_key, customer_id=self.customer_id)
         self._connection_started = False
         self._last_transcript_text = ""
-        
+
         self._register_event_handlers()
 
     def _register_event_handlers(self):
         """Registers handlers for the Bodhi client's transcription events."""
         self.client.on(LiveTranscriptionEvents.Transcript, self._on_transcript)
-        self.client.on(LiveTranscriptionEvents.UtteranceEnd, self._on_utterance_end)
-        self.client.on(LiveTranscriptionEvents.SpeechStarted, self._on_speech_started)
+        self.client.on(LiveTranscriptionEvents.UtteranceEnd,
+                       self._on_utterance_end)
+        self.client.on(LiveTranscriptionEvents.SpeechStarted,
+                       self._on_speech_started)
         self.client.on(LiveTranscriptionEvents.Error, self._on_error)
         self.client.on(LiveTranscriptionEvents.Close, self._on_close)
 
@@ -69,7 +83,8 @@ class NavanaSTT(BaseSTT):
             self._last_transcript_text = response.text
             event = STTResponse(
                 event_type=SpeechEventType.INTERIM,
-                data=SpeechData(text=response.text, language=self.language, confidence=1.0)
+                data=SpeechData(text=response.text,
+                                language=self.language, confidence=1.0)
             )
             await self._transcript_callback(event)
 
@@ -80,7 +95,8 @@ class NavanaSTT(BaseSTT):
             self._last_transcript_text = ""
             event = STTResponse(
                 event_type=SpeechEventType.FINAL,
-                data=SpeechData(text=final_text, language=self.language, confidence=1.0)
+                data=SpeechData(text=final_text,
+                                language=self.language, confidence=1.0)
             )
             await self._transcript_callback(event)
 
@@ -95,7 +111,7 @@ class NavanaSTT(BaseSTT):
     async def _on_close(self):
         print("Navana SDK connection closed.")
         self._connection_started = False
-        
+
     async def process_audio(
         self,
         audio_frames: bytes,
@@ -113,19 +129,20 @@ class NavanaSTT(BaseSTT):
                 )
                 await self.client.start_connection(config=config)
                 self._connection_started = True
-            
+
             raw_audio_data = np.frombuffer(audio_frames, dtype=np.int16)
             stereo_audio = raw_audio_data.reshape(-1, 2)
             mono_audio_float = stereo_audio.astype(np.float32).mean(axis=1)
             resampled_data = signal.resample(
-                mono_audio_float, 
-                int(len(mono_audio_float) * self.target_sample_rate / self.input_sample_rate)
+                mono_audio_float,
+                int(len(mono_audio_float) *
+                    self.target_sample_rate / self.input_sample_rate)
             )
-            
+
             audio_bytes = resampled_data.astype(np.int16).tobytes()
-            
+
             await self.client.send_audio_stream(audio_bytes)
-            
+
         except Exception as e:
             error_message = f"Audio processing error: {str(e)}"
             print(error_message)
@@ -134,8 +151,7 @@ class NavanaSTT(BaseSTT):
             if self.client._live_client and not self.client._live_client.is_closed:
                 await self.client.close_connection()
 
-
     async def aclose(self) -> None:
         """Cleans up resources by closing the SDK connection."""
         if self._connection_started:
-            await self.client.close_connection()
+            await self.client.close_connection()

videosdk_plugins_navana-0.0.32/videosdk/plugins/navana/version.py

@@ -0,0 +1 @@
+__version__ = "0.0.32"

videosdk_plugins_navana-0.0.30/videosdk/plugins/navana/version.py

@@ -1 +0,0 @@
-__version__ = "0.0.30"