Starboard Module Reference: speech_synthesis.h

  • This API provides text-to-speech functionality for audio accessibility, enabling non-visual application navigation.

  • The API includes functions to cancel all speech, check for platform support, and enqueue text for speaking.

  • The SbSpeechSynthesisSpeak function queues text to be spoken in the system's locale, playing it after any previous audio is complete.

  • API functions do not need to be thread-safe and are designed for use within a single application thread.

A basic text-to-speech API intended to be used for audio accessibility.

Implementations of this API should audibly play back text to assist users in non-visual navigation of the application.

Note that these functions do not have to be thread-safe. They must only be called from a single application thread.

Functions

SbSpeechSynthesisCancel

Cancels all speaking and queued speech synthesis audio. Must return immediately.

Declaration

void SbSpeechSynthesisCancel()

SbSpeechSynthesisIsSupported

Returns whether the platform supports speech synthesis

Declaration

bool SbSpeechSynthesisIsSupported()

SbSpeechSynthesisSpeak

Enqueues text, a UTF-8 string, to be spoken. Returns immediately.

Spoken language for the text should be the same as the locale returned by SbSystemGetLocaleId().

If audio from previous SbSpeechSynthesisSpeak() invocations is still processing, the current speaking should continue and this new text should be queued to play when the previous utterances are complete.

Declaration

void SbSpeechSynthesisSpeak(const char *text)