OpenAI.AudioGeneration.CreateRaw

OpenAI.AudioGeneration.CreateRaw

Method signatures

 Shared Function CreateRaw(Request As OpenAI.Request) As MemoryBlock
 Shared Function CreateRaw(Input As String, Model As OpenAI.Model, Voice As String, Speed As Double = 1.0, ResponseFormat As String = "mp3") As MemoryBlock

Parameters

CreateRaw(OpenAI.Request)

Name	Type	Comment
Request	Request	An OpenAI request object.

Create(FolderItem, OpenAI.Model, String, Single, String)

Name	Type	Comment
Input	String	The text to speak.
Model	Model	The TTS model to use.
Voice	String	The AI voice to use. Valid values are "alloy", "echo", "fable", "onyx", "nova", and "shimmer". Audio samples of these voices are available in the OpenAI documentation.
Speed	Single	Optional. If specified, how fast the AI should speak. Valid values are between 0.25 and 4.0.
ResponseFormat	String	Optional. If specified, the audio file format to return. Valid values are "mp3", "opus", "aac", and "flac".

Return value

Returns a MemoryBlock containing the raw audio data. The type of response depends on the ResponseFormat specified by the request.

See also

• AudioGeneration.Create