Skip to main content
Version: Next

fishjam.agent

Agent​

class Agent:

Allows for connecting to a Fishjam room as an agent peer.

Provides callbacks for receiving audio.

Attributes:

  • id: The unique identifier of the agent.
  • room_id: The ID of the room the agent is connecting to.

init​

def __init__(id: str, room_id: str, token: str, fishjam_url: str)

Creates an Agent instance.

This constructor should not be called directly. Instead, you should call :func:fishjam.FishjamClient.create_agent.

Args:

  • id: The unique identifier for the agent.
  • room_id: The ID of the room the agent will join.
  • token: The authentication token for the agent.
  • fishjam_url: The URL of the Fishjam instance.

id​

id

room_id​

room_id

connect​

def connect(self)

Connects the agent to Fishjam to start receiving messages.

Incoming messages from Fishjam will be routed to handlers defined with :func:on_track_data.

Yields:

  • AgentSession: An active session for sending media and handling events.

Raises:

  • AgentAuthError: If authentication with the Fishjam server fails.

AgentError​

class AgentError(Exception):

Base exception class for all agent exceptions.

AgentSession​

class AgentSession:

Represents an active connection session for an Agent.

init​

def __init__(
    agent: fishjam.agent.agent.Agent,
    websocket: websockets.asyncio.client.ClientConnection
)

Initializes the AgentSession.

Args:

  • agent: The Agent instance owning this session.
  • websocket: The active websocket connection.

agent​

agent

receive​

def receive(
    self
) -> AsyncIterator[AgentResponseTrackData | AgentResponseTrackImage]

Returns an async iterator over incoming messages from Fishjam.

Yields:

  • IncomingAgentMessage: The next message received from the server.

add_track​

def add_track(
    self,
    options: fishjam.agent.agent.OutgoingAudioTrackOptions
) -> fishjam.agent.agent.OutgoingTrack

Adds a track to the connected agent with the specified options.

Args:

  • options: Configuration options and metadata for the new track.

Returns:

  • OutgoingTrack: An object used to send data over the added track.

capture_image​

def capture_image(self, track_id: str)

Requests a image capture from a remote track.

The captured image will be delivered as an :class:IncomingTrackImage message through :func:receive.

Args:

  • track_id: The identifier of the track to capture an image from.

disconnect​

def disconnect(self)

Ends the agent session by closing the websocket connection.

Useful when you don't use the context manager to obtain the session.

AgentAuthError​

class AgentAuthError(fishjam.agent.errors.AgentError):

Agent failed to authenticate properly.

init​

def __init__(reason: str)

reason​

reason

IncomingTrackData​

IncomingTrackData = <class 'AgentResponseTrackData'>

IncomingTrackImage​

IncomingTrackImage = <class 'AgentResponseTrackImage'>

OutgoingTrack​

class OutgoingTrack:

Represents an outgoing track of an agent connected to Fishjam.

This is created by :func:Agent.add_track.

Attributes:

  • id: The global identifier of the track.
  • session: The agent the track belongs to.
  • options: The parameters used to create the track.

init​

def __init__(
    id: str,
    session: fishjam.agent.agent.AgentSession,
    options: fishjam.agent.agent.OutgoingAudioTrackOptions
)

id​

id: str

session​

session: fishjam.agent.agent.AgentSession

options​

options: fishjam.agent.agent.OutgoingAudioTrackOptions

send_chunk​

def send_chunk(self, data: bytes)

Sends a chunk of audio to Fishjam on this track.

Peers connected to the room of the agent will receive this data.

Args:

  • data: The raw audio bytes to send.

interrupt​

def interrupt(self)

Interrupts the current track.

Any audio that has been sent, but not played, will be cleared and prevented from playing. Audio sent after the interrupt will be played normally.

OutgoingAudioTrackOptions​

class OutgoingAudioTrackOptions:

Parameters of an outgoing audio track.

Attributes:

  • encoding: The encoding of the audio source. Defaults to raw 16-bit PCM.
  • sample_rate: The sample rate of the audio source. Defaults to 16000.
  • channels: The number of channels in the audio source. Supported values are 1 (mono) and 2 (stereo). Defaults to 1 (mono).
  • metadata: Custom metadata for the track. Must be JSON-encodable.

init​

def __init__(
    encoding: TrackEncoding = <TrackEncoding.TRACK_ENCODING_UNSPECIFIED: 0>,
    sample_rate: Literal[16000, 24000] = 16000,
    channels: Literal[1, 2] = 1,
    metadata: dict[str, typing.Any] | None = None
)

encoding​

encoding: TrackEncoding = <TrackEncoding.TRACK_ENCODING_UNSPECIFIED: 0>

sample_rate​

sample_rate: Literal[16000, 24000] = 16000

channels​

channels: Literal[1, 2] = 1

metadata​

metadata: dict[str, typing.Any] | None = None