Gemini Enterprise for Customer Experience v1beta API - Class SessionInput (1.0.0-beta10)

public sealed class SessionInput : IMessage<SessionInput>, IEquatable<SessionInput>, IDeepCloneable<SessionInput>, IBufferMessage, IMessage

Reference documentation and code samples for the Gemini Enterprise for Customer Experience v1beta API class SessionInput.

Input for the session.

Inheritance

object > SessionInput

Namespace

Google.Cloud.Ces.V1Beta

Assembly

Google.Cloud.Ces.V1Beta.dll

Constructors

SessionInput()

public SessionInput()

SessionInput(SessionInput)

public SessionInput(SessionInput other)
Parameter
Name Description
other SessionInput

Properties

Audio

public ByteString Audio { get; set; }

Optional. Audio data from the end user.

Property Value
Type Description
ByteString

Blob

public Blob Blob { get; set; }

Optional. Blob data from the end user.

Property Value
Type Description
Blob

Dtmf

public string Dtmf { get; set; }

Optional. DTMF digits from the end user.

Property Value
Type Description
string

Event

public Event Event { get; set; }

Optional. Event input.

Property Value
Type Description
Event

HasAudio

public bool HasAudio { get; }

Gets whether the "audio" field is set

Property Value
Type Description
bool

HasDtmf

public bool HasDtmf { get; }

Gets whether the "dtmf" field is set

Property Value
Type Description
bool

HasText

public bool HasText { get; }

Gets whether the "text" field is set

Property Value
Type Description
bool

Image

public Image Image { get; set; }

Optional. Image data from the end user.

Property Value
Type Description
Image

InputTypeCase

public SessionInput.InputTypeOneofCase InputTypeCase { get; }
Property Value
Type Description
SessionInputInputTypeOneofCase

RecordingControl

public SessionInput.Types.RecordingControl RecordingControl { get; set; }

Optional. Whether the conversation may be recorded and logged from here on.

Use this when the app, rather than the agent, decides what may be kept: for example when the end user has to consent to being recorded before anything about the call can be stored.

The value stays in effect until the client sends a different one, so it only needs to be sent when the answer changes. Whatever is in effect when a turn completes decides that turn, both whether its audio reaches the recording bucket and whether its transcript reaches conversation history, BigQuery and Cloud Logging. It covers the agent's side of the turn as well as the end user's.

Turns that complete before the client first sends this are recorded according to the app's logging settings, as though the field did not exist.

If more than one input in the same request sets it, the last one wins.

Property Value
Type Description
SessionInputTypesRecordingControl

Text

public string Text { get; set; }

Optional. Text data from the end user.

Property Value
Type Description
string

ToolResponses

public ToolResponses ToolResponses { get; set; }

Optional. Execution results for the tool calls from the client.

Property Value
Type Description
ToolResponses

Variables

public Struct Variables { get; set; }

Optional. Contextual variables for the session, keyed by name. Only variables declared in the app will be used by the CES agent.

Unrecognized variables will still be sent to the [Dialogflow agent][Agent.RemoteDialogflowAgent] as additional session parameters.

Property Value
Type Description
Struct

WillContinue

public bool WillContinue { get; set; }

Optional. A flag to indicate if the current message is a fragment of a larger input in the bidi streaming session.

When set to true, the agent defers processing until it receives a subsequent message where will_continue is false, or until the system detects an endpoint in the audio input.

NOTE: This field does not apply to audio and DTMF inputs, as they are always processed automatically based on the endpointing signal.

Property Value
Type Description
bool