SpeechToTextResponseUpdate Class

Definition

public ref class SpeechToTextResponseUpdate

[System.Diagnostics.CodeAnalysis.Experimental("MEAI001")]
public class SpeechToTextResponseUpdate

public class SpeechToTextResponseUpdate

[<System.Diagnostics.CodeAnalysis.Experimental("MEAI001")>]
type SpeechToTextResponseUpdate = class

type SpeechToTextResponseUpdate = class

Public Class SpeechToTextResponseUpdate

Inheritance

Object

SpeechToTextResponseUpdate

Attributes

ExperimentalAttribute

Remarks

SpeechToTextResponseUpdate is so named because it represents streaming updates to an speech to text generation. As such, it is considered erroneous for multiple updates that are part of the same audio speech to contain competing values. For example, some updates that are part of the same audio speech may have a null value, and others may have a non-null value, but all of those with a non-null value must have the same value (e.g. ResponseId).

The relationship between SpeechToTextResponse and SpeechToTextResponseUpdate is codified in the ToSpeechToTextResponseAsync(IAsyncEnumerable<SpeechToTextResponseUpdate>, CancellationToken) and ToSpeechToTextResponseUpdates(), which enable bidirectional conversions between the two. Note, however, that the conversion may be slightly lossy, for example if multiple updates all have different RawRepresentation objects whereas there's only one slot for such an object available in RawRepresentation.