solana-agent 24.1.0__py3-none-any.whl → 24.1.1__py3-none-any.whl

solana_agent/adapters/llm_adapter.py

@@ -3,14 +3,10 @@ LLM provider adapters for the Solana Agent system.
 
 These adapters implement the LLMProvider interface for different LLM services.
 """
-import asyncio
-import json
-from typing import Any, AsyncGenerator, Callable, Dict, Literal, Optional, Type, TypeVar
+from typing import AsyncGenerator, Literal, Type, TypeVar
 
-import httpx
 from openai import AsyncOpenAI
 from pydantic import BaseModel
-import websockets
 
 from solana_agent.interfaces.providers.llm import LLMProvider
 
@@ -160,173 +156,3 @@ class OpenAIAdapter(LLMProvider):
             return completion.choices[0].message.parsed
         except Exception as e:
             print(f"Error with beta.parse method: {e}")
-
-    async def create_realtime_session(
-        self,
-        model: str = "gpt-4o-mini-realtime-preview",
-        modalities: list = ["audio", "text"],
-        instructions: str = "You are a helpful assistant.",
-        voice: str = "alloy",
-        input_audio_format: str = "pcm16",
-        output_audio_format: str = "pcm16",
-    ) -> Dict[str, Any]:  # pragma: no cover
-        """Create a realtime session token for WebSocket communication."""
-        try:
-            # Get the API key from the AsyncOpenAI client
-            api_key = self.client.api_key
-
-            # Create an async HTTP client
-            async with httpx.AsyncClient() as client:
-                response = await client.post(
-                    "https://api.openai.com/v1/realtime/sessions",
-                    json={
-                        "model": model,
-                        "modalities": modalities,
-                        "instructions": instructions,
-                        "voice": voice,
-                        "input_audio_format": input_audio_format,
-                        "output_audio_format": output_audio_format,
-                    },
-                    headers={
-                        "Authorization": f"Bearer {api_key}",
-                        "Content-Type": "application/json",
-                        "OpenAI-Beta": "realtime=v1"
-                    }
-                )
-
-                if response.status_code == 200:
-                    return response.json()
-                else:
-                    raise Exception(
-                        f"Failed to create realtime session: {response.text}")
-        except Exception as e:
-            print(f"Error creating realtime session: {str(e)}")
-            raise
-
-    async def realtime_audio_transcription(
-        self,
-        audio_generator: AsyncGenerator[bytes, None],
-        transcription_config: Optional[Dict[str, Any]] = None,
-        on_event: Optional[Callable[[Dict[str, Any]], Any]] = None,
-    ) -> AsyncGenerator[str, None]:  # pragma: no cover
-        """Stream real-time audio transcription using the Realtime API.
-
-        Args:
-            audio_generator: Async generator that yields audio chunks
-            transcription_config: Optional custom configuration for transcription
-            on_event: Optional callback function for handling raw events
-
-        Yields:
-            Transcription text as it becomes available
-        """
-        # Create default transcription config if none provided
-        if transcription_config is None:
-            transcription_config = {
-                "input_audio_format": "pcm16",
-                "input_audio_transcription": {
-                    "model": "gpt-4o-mini-transcribe"
-                },
-                "turn_detection": {
-                    "type": "server_vad",
-                    "threshold": 0.5,
-                    "prefix_padding_ms": 300,
-                    "silence_duration_ms": 200
-                }
-            }
-
-        try:
-            # Get the API key from the AsyncOpenAI client
-            api_key = self.client.api_key
-
-            # Create transcription session
-            async with httpx.AsyncClient() as client:
-                response = await client.post(
-                    "https://api.openai.com/v1/realtime/transcription_sessions",
-                    json=transcription_config,
-                    headers={
-                        "Authorization": f"Bearer {api_key}",
-                        "Content-Type": "application/json",
-                        "OpenAI-Beta": "realtime=v1"
-                    }
-                )
-
-                if response.status_code != 200:
-                    raise Exception(
-                        f"Failed to create transcription session: {response.text}")
-
-                session = response.json()
-                client_secret = session["client_secret"]["value"]
-
-            # Connect to WebSocket with proper headers as dictionary
-            url = "wss://api.openai.com/v1/realtime?model=gpt-4o-mini-transcribe"
-            headers = {
-                "Authorization": f"Bearer {client_secret}",
-                "OpenAI-Beta": "realtime=v1"
-            }
-
-            async with websockets.connect(url, additional_headers=headers) as websocket:
-                # Handle WebSocket communication in the background
-                audio_task = None
-
-                async def send_audio():
-                    try:
-                        async for audio_chunk in audio_generator:
-                            # Base64 encode the audio
-                            import base64
-                            encoded_audio = base64.b64encode(
-                                audio_chunk).decode('utf-8')
-
-                            # Send audio chunk
-                            await websocket.send(json.dumps({
-                                "type": "input_audio_buffer.append",
-                                "audio": encoded_audio
-                            }))
-
-                            # Small delay to prevent flooding
-                            await asyncio.sleep(0.05)
-
-                        # Commit the audio buffer when done
-                        await websocket.send(json.dumps({
-                            "type": "input_audio_buffer.commit"
-                        }))
-                    except Exception as e:
-                        print(f"Error sending audio: {str(e)}")
-
-                # Start sending audio in the background
-                audio_task = asyncio.create_task(send_audio())
-
-                # Process transcription events
-                try:
-                    while True:
-                        message = await websocket.recv()
-                        event = json.loads(message)
-
-                        if on_event:
-                            # Check if on_event is a coroutine function and await it if needed
-                            if asyncio.iscoroutinefunction(on_event):
-                                await on_event(event)
-                            else:
-                                on_event(event)
-
-                        # Extract transcription deltas
-                        if event["type"] == "conversation.item.input_audio_transcription.delta":
-                            yield event["delta"]
-
-                        # Also handle completed transcriptions
-                        elif event["type"] == "conversation.item.input_audio_transcription.completed":
-                            yield event["transcript"]
-                            break
-                finally:
-                    # Clean up audio task if it's still running
-                    if audio_task and not audio_task.done():
-                        audio_task.cancel()
-                        try:
-                            await audio_task
-                        except asyncio.CancelledError:
-                            pass
-
-        except Exception as e:
-            print(f"Error in realtime audio transcription: {str(e)}")
-            import traceback
-            print(traceback.format_exc())
-            yield f"I apologize, but I encountered an error transcribing the audio: {str(e)}"

solana_agent/client/solana_agent.py

@@ -55,7 +55,6 @@ class SolanaAgent(SolanaAgentInterface):
         audio_input_format: Literal[
             "flac", "mp3", "mp4", "mpeg", "mpga", "m4a", "ogg", "wav", "webm"
         ] = "mp4",
-        audio_transcription_real_time: bool = True,
         router: Optional[RoutingInterface] = None,
     ) -> AsyncGenerator[Union[str, bytes], None]:  # pragma: no cover
         """Process a user message and return the response stream.
@@ -69,7 +68,6 @@ class SolanaAgent(SolanaAgentInterface):
             audio_instructions: Audio voice instructions
             audio_output_format: Audio output format
             audio_input_format: Audio input format
-            audio_transcription_real_time: Flag for real-time audio transcription
             router: Optional routing service for processing
 
         Returns:
@@ -85,7 +83,6 @@ class SolanaAgent(SolanaAgentInterface):
             audio_input_format=audio_input_format,
             prompt=prompt,
             router=router,
-            audio_transcription_real_time=audio_transcription_real_time,
         ):
             yield chunk
 

solana_agent/interfaces/client/client.py

@@ -24,7 +24,6 @@ class SolanaAgent(ABC):
             "flac", "mp3", "mp4", "mpeg", "mpga", "m4a", "ogg", "wav", "webm"
         ] = "mp4",
         router: Optional[RoutingInterface] = None,
-        audio_transcription_real_time: bool = True,
     ) -> AsyncGenerator[Union[str, bytes], None]:
         """Process a user message and return the response stream."""
         pass

solana_agent/interfaces/services/agent.py

@@ -34,7 +34,6 @@ class AgentService(ABC):
             "flac", "mp3", "mp4", "mpeg", "mpga", "m4a", "ogg", "wav", "webm"
         ] = "mp4",
         prompt: Optional[str] = None,
-        audio_transcription_real_time: bool = True,
     ) -> AsyncGenerator[Union[str, bytes], None]:
         """Generate a response from an agent."""
         pass

solana_agent/interfaces/services/query.py

@@ -23,7 +23,6 @@ class QueryService(ABC):
         ] = "mp4",
         prompt: Optional[str] = None,
         router: Optional[RoutingInterface] = None,
-        audio_transcription_real_time: bool = True,
     ) -> AsyncGenerator[Union[str, bytes], None]:
         """Process the user request and generate a response."""
         pass

solana_agent/services/agent.py

@@ -176,7 +176,6 @@ class AgentService(AgentServiceInterface):
         audio_input_format: Literal[
             "flac", "mp3", "mp4", "mpeg", "mpga", "m4a", "ogg", "wav", "webm"
         ] = "mp4",
-        audio_transcription_real_time: bool = True,
         prompt: Optional[str] = None,
     ) -> AsyncGenerator[Union[str, bytes], None]:  # pragma: no cover
         """Generate a response with support for text/audio input/output."""
@@ -195,22 +194,8 @@ class AgentService(AgentServiceInterface):
             # Handle audio input if provided - KEEP REAL-TIME AUDIO TRANSCRIPTION
             query_text = ""
             if not isinstance(query, str):
-                if audio_transcription_real_time and hasattr(self.llm_provider, "realtime_audio_transcription"):
-                    # Use realtime transcription for faster processing if available
-                    print("Using realtime audio transcription")
-                    async for transcript in self.llm_provider.realtime_audio_transcription(
-                        audio_generator=self._bytes_to_generator(query),
-                        transcription_config={
-                            "input_audio_format": audio_input_format}
-                    ):
-                        query_text += transcript
-                else:
-                    # Fall back to standard transcription
-                    print("Using standard audio transcription")
-                    async for transcript in self.llm_provider.transcribe_audio(query, input_format=audio_input_format):
-                        query_text += transcript
-
-                print(f"Transcribed query: {query_text}")
+                async for transcript in self.llm_provider.transcribe_audio(query, input_format=audio_input_format):
+                    query_text += transcript
             else:
                 query_text = query
 

solana_agent/services/query.py

@@ -47,7 +47,6 @@ class QueryService(QueryServiceInterface):
         audio_input_format: Literal[
             "flac", "mp3", "mp4", "mpeg", "mpga", "m4a", "ogg", "wav", "webm"
         ] = "mp4",
-        audio_transcription_real_time: bool = True,
         prompt: Optional[str] = None,
         router: Optional[RoutingServiceInterface] = None,
     ) -> AsyncGenerator[Union[str, bytes], None]:  # pragma: no cover
@@ -61,7 +60,6 @@ class QueryService(QueryServiceInterface):
             audio_instructions: Audio voice instructions
             audio_output_format: Audio output format
             audio_input_format: Audio input format
-            audio_transcription_real_time: Flag for real-time audio transcription
             prompt: Optional prompt for the agent
             router: Optional routing service for processing
 
@@ -122,7 +120,6 @@ class QueryService(QueryServiceInterface):
                     audio_output_format=audio_output_format,
                     audio_instructions=audio_instructions,
                     prompt=prompt,
-                    audio_transcription_real_time=audio_transcription_real_time,
                 ):
                     yield audio_chunk
 
@@ -141,7 +138,6 @@ class QueryService(QueryServiceInterface):
                     memory_context=memory_context,
                     output_format="text",
                     prompt=prompt,
-                    audio_transcription_real_time=audio_transcription_real_time,
                 ):
                     yield chunk
                     full_text_response += chunk

{solana_agent-24.1.0.dist-info → solana_agent-24.1.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: solana-agent
-Version: 24.1.0
+Version: 24.1.1
 Summary: Agentic IQ
 License: MIT
 Keywords: ai,openai,ai agents,agi
@@ -14,11 +14,9 @@ Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Dist: httpx (>=0.28.1,<0.29.0)
 Requires-Dist: openai (>=1.71.0,<2.0.0)
 Requires-Dist: pydantic (>=2.11.2,<3.0.0)
 Requires-Dist: pymongo (>=4.11.3,<5.0.0)
-Requires-Dist: websockets (>=15.0.1,<16.0.0)
 Requires-Dist: zep-cloud (>=2.9.0,<3.0.0)
 Project-URL: Documentation, https://docs.solana-agent.com
 Project-URL: Repository, https://github.com/truemagic-coder/solana-agent
@@ -375,15 +373,6 @@ async for response in solana_agent.process("user123", audio_content, output_form
     print(response, end="")
 ```
 
-### Real-Time Audio Transcription
-
-It is possible to disable real-time audio transcription responses to save on costs.
-
-```python
-async for response in solana_agent.process("user123", "What is the latest news on Canada?", audio_transcription_real_time=False):
-    print(response, end="")
-```
-
 ## Tools
 
 Tools can be used from plugins like Solana Agent Kit (sakit) or via inline tools. Tools available via plugins integrate automatically with Solana Agent.

{solana_agent-24.1.0.dist-info → solana_agent-24.1.1.dist-info}/RECORD

@@ -1,22 +1,22 @@
 solana_agent/__init__.py,sha256=ceYeUpjIitpln8YK1r0JVJU8mzG6cRPYu-HLny3d-Tw,887
 solana_agent/adapters/__init__.py,sha256=tiEEuuy0NF3ngc_tGEcRTt71zVI58v3dYY9RvMrF2Cg,204
-solana_agent/adapters/llm_adapter.py,sha256=LLRRIhtJcPrNd2qIAHmEsFE5YyuUg53-POoiNKIradQ,12833
+solana_agent/adapters/llm_adapter.py,sha256=J4Ftsbk8Yx1cEYQXhaf7WXYDBCoA4FsE6pxOeFGvcHo,5573
 solana_agent/adapters/mongodb_adapter.py,sha256=qqEFbY_v1XGyFXBmwd5HSXSSHnA9wWo-Hm1vGEyIG0k,2718
 solana_agent/client/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-solana_agent/client/solana_agent.py,sha256=iIRuwOP1jChAgiP_ewW2lEOV-PE6AtVROlt-s8mBbyg,5415
+solana_agent/client/solana_agent.py,sha256=M2AHloEFXEAM321je9xRdos5dXNQigQ0uYqnzXv7-iA,5208
 solana_agent/domains/__init__.py,sha256=HiC94wVPRy-QDJSSRywCRrhrFfTBeHjfi5z-QfZv46U,168
 solana_agent/domains/agent.py,sha256=WTo-pEc66V6D_35cpDE-kTsw1SJM-dtylPZ7em5em7Q,2659
 solana_agent/domains/routing.py,sha256=UDlgTjUoC9xIBVYu_dnf9-KG_bBgdEXAv_UtDOrYo0w,650
 solana_agent/factories/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 solana_agent/factories/agent_factory.py,sha256=mJQb1G0-gebizZvSVHm4NAxRMB1kemm2w_BAcYlN15Y,5496
 solana_agent/interfaces/__init__.py,sha256=IQs1WIM1FeKP1-kY2FEfyhol_dB-I-VAe2rD6jrVF6k,355
-solana_agent/interfaces/client/client.py,sha256=ymZiJEVy966HKVTZR75MdcrTfct6MPielHKdvfCYF_g,1742
+solana_agent/interfaces/client/client.py,sha256=CB8YuSsn-Lvinrb12huyIVaFpJqVDh8EHsHJi9SVXM4,1690
 solana_agent/interfaces/plugins/plugins.py,sha256=T8HPBsekmzVwfU_Rizp-vtzAeYkMlKMYD7U9d0Wjq9c,3338
 solana_agent/interfaces/providers/data_storage.py,sha256=NqGeFvAzhz9rr-liLPRNCGjooB2EIhe-EVsMmX__b0M,1658
 solana_agent/interfaces/providers/llm.py,sha256=09E6NgMcIpf_nJGgdVLjlZAF2HGHtW5EmhIbaEiylt0,1972
 solana_agent/interfaces/providers/memory.py,sha256=oNOH8WZXVW8assDigIWZAWiwkxbpDiKupxA2RB6tQvQ,1010
-solana_agent/interfaces/services/agent.py,sha256=KHGFjmxj0yE04VTeNa6Jpk-34OEMhDgAtzmPkpUBdRA,2165
-solana_agent/interfaces/services/query.py,sha256=2i-Qq4Bel5P5U1O5wWUYzYoECFwiMkNj7n0K1v1edd4,1532
+solana_agent/interfaces/services/agent.py,sha256=ETAfz_VbtOgpTDIpo9tMSJnUAM5boPJXw9R7b_WEu3o,2113
+solana_agent/interfaces/services/query.py,sha256=yo2JZPJSy2iwxtkUlMz0emm9u_27xNgnrAFJRHQoulQ,1480
 solana_agent/interfaces/services/routing.py,sha256=UzJC-z-Q9puTWPFGEo2_CAhIxuxP5IRnze7S66NSrsI,397
 solana_agent/plugins/__init__.py,sha256=coZdgJKq1ExOaj6qB810i3rEhbjdVlrkN76ozt_Ojgo,193
 solana_agent/plugins/manager.py,sha256=Il49hXeqvu0b02pURNNp7mY8kp9_sqpi_vJIWBW5Hc0,5044
@@ -26,10 +26,10 @@ solana_agent/plugins/tools/auto_tool.py,sha256=DgES_cZ6xKSf_HJpFINpvJxrjVlk5oeqa
 solana_agent/repositories/__init__.py,sha256=fP83w83CGzXLnSdq-C5wbw9EhWTYtqE2lQTgp46-X_4,163
 solana_agent/repositories/memory.py,sha256=75zuqAMn4YFafiLsE8RvjFNd3p5ensXbFWv6VvlhFtE,7297
 solana_agent/services/__init__.py,sha256=ab_NXJmwYUCmCrCzuTlZ47bJZINW0Y0F5jfQ9OovidU,163
-solana_agent/services/agent.py,sha256=d6Sv6W6Vtuhf5JHknUchjAD8XSUOkXALkIImnre93j8,25524
-solana_agent/services/query.py,sha256=vWopHKES-K0KpxPCSZNyunRJrkBVGGQC13awd0Sd56M,11450
+solana_agent/services/agent.py,sha256=9hL3j6efi8XfA3Xn7X1UB1eaQ1Ic_Og2mrYqjetHtYI,24646
+solana_agent/services/query.py,sha256=IFEWYfkDCbp8W0FDooAor_UZe7H1cqgrud-CzoGlu-8,11154
 solana_agent/services/routing.py,sha256=PMCSG5m3uLMaHMj3dxNvNfcFZaeaDi7kMr7AEBCzwDE,6499
-solana_agent-24.1.0.dist-info/LICENSE,sha256=BnSRc-NSFuyF2s496l_4EyrwAP6YimvxWcjPiJ0J7g4,1057
-solana_agent-24.1.0.dist-info/METADATA,sha256=WC9LoaQVgFHhA0bfXC_c57iYU7V-ZW1TJVDrCbmUmh0,20685
-solana_agent-24.1.0.dist-info/WHEEL,sha256=fGIA9gx4Qxk2KDKeNJCbOEwSrmLtjWCwzBz351GyrPQ,88
-solana_agent-24.1.0.dist-info/RECORD,,
+solana_agent-24.1.1.dist-info/LICENSE,sha256=BnSRc-NSFuyF2s496l_4EyrwAP6YimvxWcjPiJ0J7g4,1057
+solana_agent-24.1.1.dist-info/METADATA,sha256=zUuJR_cjSUaG5lSi1hDtI4veShusbIrqzGlUYMxXRmw,20307
+solana_agent-24.1.1.dist-info/WHEEL,sha256=fGIA9gx4Qxk2KDKeNJCbOEwSrmLtjWCwzBz351GyrPQ,88
+solana_agent-24.1.1.dist-info/RECORD,,