使用Google Assistant SDK时能否启用Voice Match功能？

Hey there, let's cut to the chase—Voice Match speaker recognition isn't supported by the Google Assistant SDK, regardless of whether you're running it on Windows or any non-Google-official device.

Here's the breakdown:

• Voice Match is a feature built specifically for Google's first-party hardware (like Nest Home devices, Pixel phones, etc.). It's tied to Google's cloud-based user profile verification system, which isn't exposed through the Assistant SDK for third-party or custom setups.
• The SDK gives you access to core Assistant functionality (voice commands, queries, etc.), but it doesn't include APIs for speaker authentication or linking voice profiles to Google accounts.

Are there any workarounds? #

If you need speaker recognition for your setup, you'll have to build a custom solution alongside the SDK:

• Use third-party voice biometric libraries (like Python-based tools for speaker verification) to handle local voice profile matching. You'd first record a sample of your voice to create a profile, then compare incoming audio against that profile before sending it to the Assistant SDK for processing.
• Keep in mind this won't integrate with Google's official Voice Match features—you won't get access to account-specific personalization (like pulling your calendar or reminders) tied to the voice profile, since the SDK authenticates via your project credentials, not a user's voice-linked account.

One quick side note: If you're just looking for the "Hey Google" wake word trigger, the SDK does have a hotword detection tool (googlesamples-assistant-hotword) you can use. But this is only for waking the Assistant, not verifying who's speaking.

内容的提问来源于stack exchange，提问作者Akiyoshi Okano