Module speech 

Gemini text-to-speech model (generateContent with AUDIO modality).

Structs

GoogleSpeechModel

Text-to-speech model backed by the Gemini TTS models.

GoogleSpeechOptions

Speech options (provider_options["google"]).

PreparedSpeechRequest

A prepared speech request.

Constants

DEFAULT_SAMPLE_RATE

Sample rate assumed when the response media type carries none.

DEFAULT_VOICE

Voice used when none is requested.

FAMILY

Provider id family.

Functions

add_wav_header

Wraps signed 16-bit little-endian mono PCM in a 44-byte WAV header.

parse_sample_rate

Sample rate encoded in a media type such as audio/L16;codec=pcm;rate=24000.