Google Cloud Text-To-Speech被误收WaveNet费用问题咨询

Hey there! Let's break down why you're getting charged for WaveNet and how to switch away from it.

In your current code, you only specify the LanguageCode and SsmlGender in the VoiceSelectionParams—you aren’t picking a specific voice name or model type. Google Text-to-Speech automatically selects a default voice that matches your criteria, and for en-US female voices, the default is often a WaveNet model (like en-US-Wavenet-F). WaveNet is a premium paid model, which is why those charges are appearing on your bill.

You have two straightforward ways to avoid WaveNet fees:

1. Specify a Standard Model Voice Name #

Add a VoiceName parameter to explicitly choose a non-WaveNet voice. For example, en-US-Standard-F is a standard (free-tier eligible) female voice for English-US:

Voice = new VoiceSelectionParams
{
    LanguageCode = "en-US",
    SsmlGender = SsmlVoiceGender.Female,
    VoiceName = "en-US-Standard-F" // Use a standard, non-WaveNet voice
}

All standard model voices follow the naming pattern [Language-Code]-Standard-[Gender-Identifier], so you can pick others from Google’s TTS voice list if you prefer a different option.

2. Force the Standard Model Directly #

If you don’t want to lock into a specific voice, you can set the Model field to "standard" to ensure Google uses only the free standard model, regardless of the default voice selected:

Voice = new VoiceSelectionParams
{
    LanguageCode = "en-US",
    SsmlGender = SsmlVoiceGender.Female,
    Model = "standard" // Enforce use of the free standard TTS model
}

After updating your code, you can verify the change by checking your Google Cloud Console's Text-to-Speech usage reports—you should no longer see WaveNet listed in the call details.

内容的提问来源于stack exchange，提问作者RG Servers