Azure API 기반 Text-to-Speech(TTS)의 상업적 이용

Hello @Jimin Jeong(정지민) ,

Thank you for reaching out to Microsoft support!

For your scenario—generating Text-to-Speech audio through Azure AI Speech and embedding or distributing that audio as part of commercial hardware/software products—there are a few points to consider.

1. Commercial use and distribution of generated audio

For your cloud-based audio-generation workflow, I recommend using a paid Speech resource such as Standard S0. Published Microsoft Q&A guidance supports commercial use and redistribution of audio generated using paid-tier prebuilt neural voices:

Accepted Microsoft Q&A answer on paid-tier prebuilt-voice redistribution

However, using S0 does not by itself constitute blanket legal clearance for every distribution scenario. Your applicable Microsoft agreement and the specific voice type and use case still need to be considered.

You should also ensure that:

You have the necessary rights to the input text and any other content incorporated into the generated output.

Clearly inform users that the voice/audio is AI-generated or synthetic, in accordance with the Microsoft Enterprise AI Services Code of Conduct.

If you are using a custom neural voice or personal voice, additional requirements and restrictions may apply, and the guidance for standard/prebuilt neural voices should not automatically be applied to those voice types.

2. Recommended plan/tier

For generating audio files through the Azure cloud service and packaging those files with your product, Standard S0 is an appropriate starting point.

The appropriate Speech tier should also be selected based on your expected character volume, throughput, latency, and other technical requirements. For the cloud-based workflow described here, the paid Speech tier should be evaluated based on those workload requirements rather than treating the tier itself as a separate redistribution license.

If your requirement is instead to run the speech-synthesis engine locally/on-device, that is a different Azure Speech offering with separate availability and requirements. See the Microsoft documentation for Embedded Speech.

3. Azure API key and Text-to-Speech development

Microsoft provides official documentation and sample code for developing Text-to-Speech applications using an Azure Speech resource.

For getting started:

Text-to-speech quickstart

For generating audio files specifically:

Synthesize speech to a file

The Speech REST API documentation also provides examples of authenticating requests with an API key:

Text-to-speech REST API

For production applications, protect Speech resource credentials carefully. Do not expose an API key in client-side applications or publicly distributed code. Where supported by your architecture, consider Microsoft Entra authentication/managed identities, and use a service such as Azure Key Vault for securely storing secrets and credentials.

Contract-specific confirmation

For contract-specific confirmation of your redistribution rights, I recommend asking your Microsoft account/licensing representative to review the applicable agreement, voice type, and distribution mode.

Please "Upvote the Answer" if this information helped you. This will help us and others in the community as well.