Quick start
Set up an account, services, a project, and credentials to use Intelligent Speech Interaction for speech recognition and speech synthesis through SDKs or APIs.
Procedure
The setup has six steps: prepare an account, create an AccessKey pair, activate services, manage projects, obtain a token, and integrate the service. Skip the corresponding creation steps if the account, services, or credentials are already available.
To try services only in the console, no AccessKey pair or manually obtained token is needed.
Step 1: Prepare an account
Prepare an Alibaba Cloud account. If you do not have one, see Create an Alibaba Cloud account. For an individual account, follow the instructions in Individual real-name verification.
For collaboration, use Resource Access Management (RAM) users to assign permissions instead of sharing an Alibaba Cloud account password or AccessKey pair. For instructions, see Create a RAM user.
Grant the required permissions before a RAM user calls the service. The AliyunNLSFullAccess system policy provides full access to Intelligent Speech Interaction. Grant permissions based on actual needs. For instructions, see Grant permissions to a RAM user.
Step 2: Create an AccessKey pair
An AccessKey pair authenticates calls to Alibaba Cloud APIs. It consists of an AccessKey ID, which identifies the user, and an AccessKey secret, which verifies the identity. An existing, valid AccessKey pair can be reused.
AccessKey pair for a RAM user (recommended): Sign in to the RAM console with an Alibaba Cloud account and create an AccessKey pair for an authorized RAM user. For instructions, see Create an AccessKey pair for a RAM user.
AccessKey pair for an Alibaba Cloud account: This pair provides full access to the account and is not recommended for routine development. If it is required, go to the AccessKey management page and follow the instructions in Create an AccessKey pair for an Alibaba Cloud account.
The AccessKey secret is displayed only when the pair is created. Store it securely. Do not include it in public code, logs, or client-side code.
Step 3: Activate services
Sign in to the Intelligent Speech Interaction console with an Alibaba Cloud account and activate the services to use.
|
Service edition |
Description |
|
Free trial |
Some services offer a three-month free trial with two concurrent calls or two hours of recording file recognition per day. Quotas vary by service. |
|
Commercial edition |
Charges are based on actual usage after activation. Use this edition for production workloads or services that do not offer a free trial. |
For trial quotas and pricing, see Billing. To continue using a service after its trial expires, see Upgrade to the commercial edition.
Step 4: Manage projects
Create a project and obtain its Appkey. The Appkey identifies the project used in service calls. For instructions, see Manage projects.
Project management is available in the Intelligent Speech Interaction console.
Step 5: Obtain a token
An access token authenticates calls to Intelligent Speech Interaction services. Its expiration time is specified by the ExpireTime value returned by the API. Reuse the token while it is valid. The application must obtain a new token before the current one expires; do not rely on automatic SDK refresh.
For a quick test, obtain a token in the console. For production, use one of the following methods.
|
Method |
When to use |
|
Use an AccessKey pair to obtain a token through the SDK. Manage token renewal in the application. |
|
|
No SDK is available for the programming language, or dependencies need to be managed directly. |
Step 6: Integrate the service
Configure the sample with the project's Appkey and a valid token, then call an activated service. The AccessKey pair used to obtain the token must have permission to access resources under the account that owns the project.
Choose a starting point based on whether coding is required.
|
Starting point |
Audience |
Description |
|
Developers |
Try speech recognition and speech synthesis with curl, Postman, or an SDK. |
|
|
Non-developers |
Try speech recognition and speech synthesis in the console without writing code. |
For other integration methods, see SDK and API overview.