STT Options

SttOptions configures both transcribe and the streaming calls.

let options = SttOptions(
    language: "en",
    punctuation: true,
    wordTimestamps: true,
    diarization: false,
    maxSpeakers: nil,
    translateToEnglish: false
)

let transcription = try await RunAnywhere.stt.transcribe(audio, options: options)

Fields

Field	Type	Default	Meaning
`language`	`String?`	none, auto-detect	BCP-47 code such as `"en"` or `"es"`
`punctuation`	`Bool`	proto default	Insert punctuation and casing
`wordTimestamps`	`Bool`	proto default	Per-word timings in the result
`diarization`	`Bool`	`false`	Attribute speech to speakers
`maxSpeakers`	`Int?`	none	Cap the speaker count when diarizing
`translateToEnglish`	`Bool`	`false`	Translate rather than transcribe

Defaults for punctuation and wordTimestamps come from RASTTOptions.defaults() rather than being hardcoded here, so they follow the C++ core.

Language

Leave language unset to auto-detect. Set it when you know the language: detection costs time and can pick wrong on short clips.

SttOptions(language: "es")

Word timestamps

let transcription = try await RunAnywhere.stt.transcribe(
    audio,
    options: SttOptions(wordTimestamps: true)
)

Use these to build captions or to seek a player to a spoken word.

Diarization

diarization: true labels who spoke inside the transcription. For a standalone answer to “who spoke when”, use the diarization namespace instead:

let result = try await RunAnywhere.diarization.diarize(AudioInput.file(path))

Translation

translateToEnglish: true returns English regardless of the source language, where a plain transcription returns the original.

Getting Started

Swift SDK

Kotlin SDK

React Native SDK

Flutter SDK

Web SDK

Electron SDK

Vibe Coding

Fields

Language

Word timestamps

Diarization

Translation

​Fields

​Language

​Word timestamps

​Diarization

​Translation

Fields

Language

Word timestamps

Diarization

Translation