openai-core/com.aallam.openai.api.audio/TranscriptionRequest

TranscriptionRequest

class TranscriptionRequest(val audio: FileSource, val model: ModelId, val prompt: String? = null, val responseFormat: AudioResponseFormat? = null, val temperature: Double? = null, val language: String? = null, val timestampGranularities: List<TimestampGranularity>? = null)

Request to transcribe audio into the input language.

Constructors

constructor(audio: FileSource, model: ModelId, prompt: String? = null, responseFormat: AudioResponseFormat? = null, temperature: Double? = null, language: String? = null, timestampGranularities: List<TimestampGranularity>? = null)

Properties

audio

val audio: FileSource

The audio file to transcribe, in one of these formats: mp3, mp4, mpeg, mpga, m4a, wav, or webm.

language

val language: String? = null

The language of the input audio. Supplying the input language in ISO-639-1 format will improve accuracy and latency.

model

val model: ModelId

ID of the model to use. Only whisper-1 is currently available.

prompt

val prompt: String? = null

An optional text to guide the model's style or continue a previous audio segment. The prompt should match the audio language.

responseFormat

val responseFormat: AudioResponseFormat? = null

The format of the transcript output, in one of these options: json, text, srt, verbose_json, or vtt.

temperature

val temperature: Double? = null

The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.

timestampGranularities

val timestampGranularities: List<TimestampGranularity>? = null

The timestamp granularities to populate for this transcription. responseFormat must be set AudioResponseFormat.VerboseJson to use timestamp granularities. Either or both of these options are supported: TimestampGranularity.Word, or TimestampGranularity.Segment. Note: There is no additional latency for segment timestamps, but generating word timestamps incurs additional latency.