StreamSpeechResult
Namespace: Vercel.AI.Operations
public sealed class StreamSpeechResultResult of CancellationToken). Maps to StreamSpeechResult.
Inheritance: object
Properties
Section titled “Properties”AudioStream
Section titled “AudioStream”public OperationFullStream<byte[]> AudioStream { get; }Stream of audio bytes in generation order. AudioStream and FullStream share one live stream: only one of them can be accessed, and only once. Access it before any result promise.
Value: OperationFullStream<byte\[\]>
FullStream
Section titled “FullStream”public OperationFullStream<SpeechStreamPart> FullStream { get; }Full stream of speech parts (audio, raw, and error parts). AudioStream and FullStream share one live stream: only one of them can be accessed, and only once. Access it before any result promise.
Value: OperationFullStream<SpeechStreamPart>
ProviderMetadata
Section titled “ProviderMetadata”public Task<JsonElement> ProviderMetadata { get; }Additional provider-specific metadata. Accessing it claims the stream.
Value: Task<JsonElement>
Responses
Section titled “Responses”public Task<IReadOnlyList<ProviderResponse>> Responses { get; }Response metadata. Accessing it claims the stream.
Value: Task<IReadOnlyList<ProviderResponse>>
public Task<SpeechStreamUsage> Usage { get; }Usage reported for the call, if available. Accessing it claims the stream.
Value: Task<SpeechStreamUsage>
Warnings
Section titled “Warnings”public Task<IReadOnlyList<OperationWarning>> Warnings { get; }Warnings for the call, such as unsupported settings. Accessing it claims the stream.
Apache-2.0 · Copyright 2023 Vercel, Inc. · unofficial community port, not an official Vercel product