GuidesAPI ReferenceSDKsIntegrationsWhat's New?
Guides

Streaming API

Suggest Edits

The Streaming API is based on WebSocket protocol and can be used for real-time use cases where both the audio and its insights need to be available in real-time.

📘

Streaming Feature Availability

Currently, you can get only Questions and Action Items in real-time with Streaming API for Python SDK. However, you can get Messages, Topics, Action Items, Follow-Ups, Questions, Members and Conversation data using the get Conversation object function.

The Python SDK provides the following capabilities:

Start Connection

The code snippet below allows you to a Streaming API connection with Symbl via WebSocket. This returns a connection object:

import symbl
connection_object = symbl.Streaming.start_connection(
    insight_types=["question", "action_item"],
    speaker={"name": "John", "userId": "[email protected]"},
)

# print(connection_object)
ParameterRequiredDescription
credentialsOptionalDon't add this parameter if you have symbl.conf file in your home directory or working directory. Read more about symbl.conf file and using credentials in the Python SDK Configuration section.
speakerOptionalSpeaker object containing name and userId field.
insight_typesOptionalThe insights to be returned in the WebSocket connection. Questions and Action Items are the insights currently supported.
configOptionalUse this parameter to pass confidenceThreshold, languageCode For more details, see the config parameter documentation at Streaming API.

Stop Connection

To stop an active WebSocket connection, use the code given below:

import symbl

connection_object.stop()

Subscribe to Events

Once the WebSocket connection is established, you can get live updates on conversation events such as generation of transcript, action items or questions, etc.

import symbl

connection_object.subscribe(
    {
        "message_response": lambda response: print(
            "got this response from callback", response
        ),
        "message": lambda response: print("got this response from callback", response),
        "topic_response": lambda response: print(
            "got this response from callback", response
        ),
    }
)

The connection_object.subscribe is a function of the connection object that listens to the events of a live call and let's you subscribe to them in real-time. It takes a dictionary parameter, where the key can be an event and its value can be a callback function that should be executed on the occurrence of that event.

Supported Events

Following are the functions for different events you can subscribe to:

EventDescription
message_responseGenerates an event whenever transcription is available.
messageIt will generate an event for live transcriptions. It will include isFinal property which will be False initially, meaning the transcription is not finalized.
insight_responseGenerates an event whenever an action_item or question is identified in the message.
topic_responseGenerates an event whenever a topic is identified in any transcription.

Send Audio from Mic

This allows you to send data to WebSocket directly via your mic. It is a recommended function for first time users sending audio to Symbl.

To receive the insights via email, use the code given below:

import symbl
connection_object.send_audio_from_mic()

Send Audio Data

You can send custom audio data from some other library using the following code.

The send_audio function sends audio data to WebSockets in binary format.

import symbl
connection_object.send_audio(data)

Get Conversation Object

Connection object has a conversation parameter through which you can directly query the Conversation API with the provided conversationId.

import symbl

connection_object = symbl.Streaming.start_connection()

...

print (connection_object.conversation.get_topics())
#print(connection_object.conversation.get_messages())
#print (conversation_object.conversation.get_action_items())
#print (conversation_object.conversation.get_follow_ups())
#print (conversation_object.conversation.get_questions())
#print (conversation_object.conversation.get_members())
#print(connection_object.conversation.get_conversation()) #Gets the conversation data from the conversation

Complete Sample Code

The sample code given below shows the usage of subscribe function and other functionalities explained above. You can simply copy-paste this code snippet and run it in your Python editor to 

import symbl

events = {
    "message_response": lambda response: print(
        "Final Messages -> ",
        [message["payload"]["content"] for message in response["messages"]],
    ),
    "message": lambda response: print(
        "live transcription : {}".format(
            response["message"]["punctuated"]["transcript"]
        )
    )
    if "punctuated" in response["message"]
    else print(response),
    "insight_response": lambda response: [
        print(
            "Insights Item of type {} detected -> {}".format(
                insight["type"], insight["payload"]["content"]
            )
        )
        for insight in response["insights"]
    ],
    "topic_response": lambda response: [
        print(
            "Topic detected -> {} with root words, {}".format(
                topic["phrases"], topic["rootWords"]
            )
        )
        for topic in response["topics"]
    ],
}
connection_object = symbl.Streaming.start_connection(
    insight_types=["question", "action_item"],
    speaker={"name": "John", "email": "[email protected]"},
)
connection_object.subscribe(events)
connection_object.send_audio_from_mic()

Python SDK Reference

For a complete list of supported classes and objects in the Python SDK, see the Python SDK Reference page.

You can view more capabilities added to the Streaming API in the following sections:

Error Code

ErrorDescription
sounddevice.PortAudioError: Error opening InputStream: Internal PortAudio error [PaErrorCode -9986]PortAudio Errors on Mac Systems: If you're getting PortAudio Error, please consider updating the PortAudio library in your system. Running the following command can help.
brew install portaudio

Updated about 2 years ago