Azure AI Speech Studio

Hi @AzureUser-9588,

Thank you for reaching out to Microsoft Q&A forum!

What is the difference between Personal Voice and Custom Neural Voice Lite in Azure Speech? I am looking for a clear distinction to help me choose one over the other.

Personal Voice and Custom Neural Voice Lite in Azure Speech serve different purposes. Personal Voice allows users to create a unique AI-generated version of their own voice with just a one-minute audio sample, making it ideal for applications where individuals want to replicate their voice across various languages. However, it is limited to specific regions and primarily aimed at business customers who wish to enable their users to use their personal voice in applications. In Custom Neural Voice Lite is a project type for demonstration and evaluation purposes, requiring users to record 20-50 predefined scripts. It provides a moderate-quality voice model and is open to anyone for testing, though deploying it for commercial use requires access to the full Custom Neural Voice professional version. While Personal Voice focuses on individual voice replication, Custom Neural Voice Lite is geared toward broader testing of neural voice technology before investing in higher-quality, professional recordings.

For more info:

• What is personal voice.
• Custom neural voice lite.

Additionally, after creating a new Personal Voice in Azure Speech Studio, how can I retrieve the speakerprofileId and use that voice in my application?

To integrate a Personal Voice into your application, you'll first need to retrieve the speakerProfileId. When you create a Personal Voice in Azure Speech Studio, the speakerProfileId is generated through the user's verbal consent and a clean audio sample, which typically ranges from 5 to 90 seconds. To create a personal voice and get the speaker profile ID, use the PersonalVoices_Post operation

Once you have the speakerProfileId, you can use it to synthesize speech by sending requests that include this ID alongside your text input. Make sure your audio files are accessible via a public URL or uploaded directly, as this is part of the creation process.

For more info: Get a speaker profile ID for the personal voice.

I hope this helps. Thank you.

Hi @AzureUser-9588,

Following up to see if the given response was helpful. Thank you.