GeminiTTS speaking_rate is not exposed, although StreamingAudioConfig supports it

[dvirginz]

Hi,

I am using Pipecat with Google TTS and Gemini TTS. I want to control the speaking rate. The underlying Google client supports speaking_rate on StreamingAudioConfig, but Pipecat does not expose it.

In my own direct calls to the Google client I can do something like this:

voice_params = texttospeech.VoiceSelectionParams(**voice_params_kwargs)

# Configure audio with speaking rate (StreamingAudioConfig supports this)
audio_config = texttospeech.StreamingAudioConfig(
    speaking_rate=self.speaking_rate
)

config = texttospeech.StreamingSynthesizeConfig(
    voice=voice_params,
    streaming_audio_config=audio_config
)

yield texttospeech.StreamingSynthesizeRequest(
    streaming_config=config
)

When I use Pipecat, I do not see any way to pass speaking_rate into the Google TTS integration. So I cannot control the speaking rate through Pipecat even though the API supports it. (GoogleHttpTTSService does support it in pipecat)

I would like Pipecat to expose speaking_rate, for example as part of the TTS options or voice config, and pass it through to StreamingAudioConfig.