class

OpenAI::TranscriptionRequest

Inherits JSON::Serializable / Reference / Object

TranscriptionRequest represents a request structure for audio API

Constructors

new(pull : JSON::PullParser)

new(file : File | Path | String, model : String = "whisper-1", prompt : Nil | String = nil, response_format : OpenAI::TranscriptionRespFormat = TranscriptionRespFormat::JSON, temperature : Float64 = 0.0, language : Nil | String = nil)

Source

new(*, __pull_for_json_serializable pull : JSON::PullParser)

Source

Instance methods

build_metada(builder : HTTP::FormData::Builder)

Source

file

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

Source

file=(file : File | Path | String)

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

Source

language

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

Source

language=(language : String | Nil)

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

Source

model

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

Source

model=(model : String)

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

Source

prompt

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

Source

prompt=(prompt : String | Nil)

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

Source

response_format

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

Source

response_format=(response_format : TranscriptionRespFormat)

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

Source

temperature

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.

Source

temperature=(temperature : Float64)

Source