Intelligent Speech Interaction is built on technologies such as speech recognition, speech synthesis, and natural language understanding to create intelligent, human-like interactive experiences. It is ideal for enterprise use cases such as intelligent customer service, quality inspection, meeting transcription, and real-time captioning. This document explains how to use Intelligent Speech Interaction and outlines its workflow and operations.
Step 1: Prepare your account
-
Register an Alibaba Cloud account.
An Alibaba Cloud account tracks your resource consumption for billing and provides full access to all Alibaba Cloud products and management features. For more information, see Sign up for an Alibaba Cloud account.
-
Complete individual real-name registration.
To use Alibaba Cloud products and services, you must complete individual real-name registration. For more information, see Individual real-name registration.
-
(Optional) Create and authorize a RAM user.
If your enterprise requires multiple users to collaborate and access resources, you can create RAM users. Resource Access Management (RAM) allows you to grant least-privilege permissions as needed. This helps avoid sharing your Alibaba Cloud account password or AccessKeys with other users, reducing security risks. For more information, see Create a RAM user. For more information about RAM users, see What is Resource Access Management?.
If you use a RAM user to call Intelligent Speech Interaction, go to the console and grant the AliyunNLSFullAccess permission to the RAM user. For more information, see Manage RAM user permissions.
On the Select Permissions page, click the System Policy tab, search for NLS, and then select AliyunNLSFullAccess.
NoteWhen you create a RAM user, you must set a logon password. Otherwise, the user cannot log on to the RAM console.
Step 2: Create an AccessKey
When you call Alibaba Cloud APIs, you must use an AccessKey to authenticate your identity. An AccessKey consists of an AccessKey ID and an AccessKey Secret:
-
AccessKey ID: Used to identify a user.
-
Keep the AccessKey Secret confidential.
Create Alibaba Cloud account AccessKey
Log on to the RAM console and use your Alibaba Cloud account to create an AccessKey. For more information, see Create an AccessKey.
Create RAM user AccessKey
Log on to the RAM console with your Alibaba Cloud account to create an AccessKey for a RAM user. For more information, see Create an AccessKey.
The AccessKey Secret for a RAM user is displayed only upon creation and cannot be retrieved later. Save it immediately, for example to a local file.
In the left-side navigation pane, choose Identity Management > Users. Click the logon name of the target user to go to the details page. On the Authentication Management tab, in the User AccessKey section, click Create AccessKey.
Step 3: Enable the service
If this is your first time using Intelligent Speech Interaction, we recommend that you enable the service with your Alibaba Cloud account.
Go to the Intelligent Speech Interaction product homepage and click Enable and Purchase. On the product enablement page, select a service type, select the Service Agreement checkbox, and then click Enable Now to activate the Intelligent Speech Interaction service.
|
Type |
Description |
|
Free Trial |
All features are included in the trial by default, except for long-text speech synthesis, audio file recognition (off-peak), and audio file recognition (express). New users receive a 3-month free trial that supports a concurrency of 2 or a daily quota of 2 hours for audio file recognition. Important
During the 3-month trial period for new users, you receive a free quota for 2 hours of file transcription that resets every 24 hours. If you use up the daily quota, you can resume using the free service after the quota resets. |
|
Commercial Edition |
Select one or more speech services for commercial use. After activation, you are billed on a Pay-As-You-Go basis. Fees are deducted from your Alibaba Cloud account balance. For more information, see Billing methods. |
If error code 40000010 (Gateway:FREE_TRIAL_EXPIRED) is returned when you call a service, it is usually caused by one of the following reasons:
Reason 1: The free trial has expired. The free trial lasts 3 months. After it expires, all services (including one-sentence recognition, real-time speech recognition, and speech synthesis) become unavailable. You must upgrade to the commercial edition.
Reason 2: A commercial-only service was called. The following services are not available for free trial and require the commercial edition even during the trial period:
-
Long text speech synthesis
-
Audio file recognition (off-peak)
-
Audio file recognition (express)
-
Streaming text-to-speech (CosyVoice)
-
VoiceChat
Solution: Go to the Service Management page to check the service status and upgrade the required services to the commercial edition. For a detailed comparison, see Edition comparison.
After selecting a service edition, click Enable Now.
Step 4: Manage projects
Log on to the Intelligent Speech Interaction console and create a project to generate a corresponding Appkey. For more information, see Manage projects.
Step 5: Obtain a token
An access token is an authentication credential required to call Intelligent Speech Interaction services.
Quick start recommendation: If this is your first time, use Obtain a token from the console for the simplest setup. For production environments, use the built-in token management in the SDK.
A token has an expiration period (typically 24 hours) and must be refreshed before it expires. For production environments, use the SDK to automatically manage token retrieval and refresh.
|
Method |
Scenario |
|
For testing purposes only. |
|
|
For production environments. Pass your AccessKey ID and AccessKey Secret to the SDK to periodically obtain a token. |
|
|
If an SDK is not available for your programming language or if you need to control dependencies, you can obtain a token by using OpenAPI. |
Step 6: Integrate and develop
Make sure the Appkey, AccessKey, and token belong to the same Alibaba Cloud account (or the same RAM user). Otherwise, API calls will fail.
After completing the steps above, choose the getting started path that best fits your needs:
|
Path |
Audience |
Description |
|
Developers |
Try speech recognition and speech synthesis by using curl commands, Postman, or SDKs. Beginners: start with the "One-sentence recognition curl example" — a single command to make your first API call. |
|
|
Non-developers |
Try speech recognition and speech synthesis directly in the console, with no coding required. |
For more integration options (RESTful API, mobile client, WebSocket, etc.), see SDK and API overview.
FAQ
-
What does error code
Gateway:FREE_TRIAL_EXPIRED(40000010) mean?This error has two common causes:
Cause 1: The free trial has expired. The free trial lasts 3 months. After expiration, all trial services become unavailable. Go to the Service Management page in the Intelligent Speech Interaction console to check the trial status. If it has expired, upgrade the required services to the commercial edition.
Cause 2: A commercial-only service was called. Even during the trial period, services such as long text speech synthesis, audio file recognition (off-peak/express), streaming text-to-speech (CosyVoice), and VoiceChat require the commercial edition and are not available for free trial.
For more information, see Activate the service. For error code details, see Error codes.
-
Which services support free trial?
Services available for free trial include: audio file recognition, real-time speech recognition, one-sentence recognition, speech synthesis, speaker recognition, sound event detection, gender recognition, and language recognition. The trial lasts 3 months with a concurrency of 2. For details, see Edition comparison.
-
How do I upgrade from free trial to commercial edition?
Log on to the Intelligent Speech Interaction console. On the Service Management page, click Upgrade to Commercial Edition next to the desired service. After activation, billing is on a pay-as-you-go basis. You can also purchase resource packages to reduce costs. For more information, see Activate the service.