Quick start

Updated at:

Voice Service provides voice communication capabilities, such as voice notifications and voice captchas. This topic describes the basic concepts and billing methods of Voice Service, explains how to use the service, and answers frequently asked questions (FAQs).

Product usage

Alibaba Cloud Voice Service provides multiple product features, including voice notifications, voice captchas, Interactive Voice Response (IVR), Intelligent Speech Interaction, and intelligent voice bots. You can choose a voice product based on your scenario or business needs.

Feature

Description

Usage method

Instructions

Voice Notification/Voice Captcha

Send a voice notification or voice captcha to a specified number.

API

Use voice notifications or voice captchas using an API

Task Assistant

Use voice notifications or voice captchas using the console

Voice Interaction (IVR)

Call the voice interaction API to start an interactive voice call to a specified number. After the call connects, the system plays an audio file. Users can then interact with your business system by pressing phone keys.

API

Use voice interaction (IVR) using an API

Intelligent Speech Interaction

Lets users have voice conversations with a machine. This feature is common in scenarios such as intelligent calls, cloud customer service, intelligent quality inspection, and AI chatbots. It improves call efficiency and helps reduce labor costs with automatic dialing and intelligent responses.

Configure callback

Intelligent Voice Bot

The intelligent voice bot feature is upgraded. To use the new features, log on to the Artificial Intelligence Cloud Call Service console and activate them on the Intelligent Outbound Robot page. If you already use this feature in Voice Service, you can continue to do so. We recommend that you migrate your services as soon as possible.

How do I choose?

Before you start, it is important to understand the differences between features and modes. This will help you make better technical and cost-effective choices.

Core concept comparison: Common mode vs. Dedicated mode

Mode

Caller ID

Cost structure

Activation speed

Number concurrency

Suitable scenarios

Common mode

Randomly assigned from a public number pool

Pay-as-you-go

Fast

Shared resources, not guaranteed

  • Testing or low business volume

  • Quick launch

Dedicated mode

Apply for a dedicated number

Monthly subscription + Pay-as-you-go

Relatively slow

Dedicated (15 channels for landlines/3 channels for mobile phones)

  • Large-scale, high-frequency calls

  • Requires IVR or intelligent conversation

For more information, see Differences between common mode and dedicated mode.

Scenario-based selection

  • Scenario 1: Sending captchas, system alerts, and event notifications

    • Recommended solution: Voice Notification or Voice Captcha + Common mode

    • Advantages: Low cost (no monthly number subscription), fast activation, and low technical requirements.

    • Limitations: The caller ID is not fixed, which may affect the answer rate. Concurrency is shared and can fluctuate.

  • Scenario 2: Business calls that need to display a fixed company number

    • Recommended solution: Voice Notification + Dedicated mode

    • Advantages: Uses a dedicated number to improve the answer rate. Dedicated concurrency provides high stability.

    • Limitations: Requires applying for a number, which incurs a monthly fee. The review process is longer than for common mode.

  • Scenario 3: Interacting with users through key presses (such as checking phone bills or satisfaction surveys)

    • Recommended solution: Voice Interaction (IVR)

    • Note: This feature requires Dedicated mode.

  • Scenario 4: Having intelligent voice conversations with users (such as intelligent customer service)

    • Recommended solution: Intelligent Speech Interaction

    • Note: This feature requires Dedicated mode.

Complete first-time usage flow

image

Prerequisites: You must register an Alibaba Cloud account, complete enterprise identity verification, and activate Voice Service.

Step 1: Complete enterprise qualification verification

  • Purpose: To verify your enterprise identity, which is a prerequisite for using domestic communication services.

  • Operation: Go to the Enterprise Qualifications page and submit your valid enterprise certificates, such as a business license.

  • Estimated time: The review is usually completed within one business day.

Step 2: Apply for business scenario scripts

  • Purpose: To report your business scenarios to carriers for compliance.

  • Operation: Go to the Scenario Scripts page. Fill out and submit the application based on your business purpose, such as captchas or order notifications.

  • Key difference: The script you apply for here is only for reporting business scenarios to carriers for compliance review. It is not played back during the call. The content that the user hears is determined by the text-to-speech template or voice file that you create in the next step. Ensure that the broadcast content you create later matches the reported scenario.

Step 3: Create the actual broadcast content

  • Purpose: To create the content that users will hear during the call.

  • Operation: You can choose one of the following methods based on your needs:

    • Text-to-speech template: On the Voice Notification management page, select Text-to-Speech Template > Add Template. You can insert variables in the Template Content field. For example, You have a new order. The order number is ${order_id}.

    • Voice notification file: On the Voice Notification management page, click Import Voice File to upload a pre-recorded voice file. MP3 and WAV formats are supported.

Step 4: Prepare numbers (Dedicated mode only)

  • Purpose: To apply for and configure dedicated numbers for calling.

  • Operation: Go to the Real Number Application page and click Apply for Number.

Step 5: Initiate calls

  • Purpose: To call users and play the voice content using the console or an API.

  • Operation: See Usage methods.

Choose a usage mode

Console

API

FAQ

Video explanation