Automatic Speech Recognition

Automatic Speech Recognition (ASR) allows you to convert audio recordings into text. You can invoke APIs to recognize audio files or speech sent from a variety of sources in real time. ASR specializes in Mandarin Chinese recognition. (The core technology of the service is provided by partner iFLYTEK.)

Fees start as low as ¥2.5 per thousand API calls.


Try Now
Getting Started
Product Advantages

  • High Accuracy

    Employs advanced deep learning technologies to achieve a speech recognition accuracy of over 95%.

  • Diverse Functions

    Supports Sentence Recognition and Long Speech Recognition. Long Speech Recognition provides a standard edition and a dedicated telephone edition for call recording recognition, meeting demanding requirements in various scenarios.

  • High Stability

    Proven stability after years of experience in complex enterprise customer scenarios.

  • High Efficiency

    Provides standard RESTful APIs and various SDKs to facilitate service use and integration while reducing labor and business costs.

Application Scenarios

  • Smart Customer Service

  • Intelligent Conferencing

  • Live Subtitles

  • Human-Machine Interaction

Smart Customer Service

Smart Customer Service

Smart customer service systems using ASR can automatically recognize and understand customers' speech, and provide correct responses, reducing labor costs and ensuring service quality.

Advantages

Accurate Recognition

High speech recognition accuracy

High Speed

Fast recognition of both short and long segments of speech

Related Services

obs

Intelligent Conferencing

Intelligent Conferencing

Intelligent conference-call systems using ASR can accurately and automatically recognize participants' voiceprints and voices so that users can obtain real-time subtitles and conference records.

Advantages

Accurate Recognition

High speech recognition accuracy

Convenience and Efficiency

Fast conference recording and quick subtitle launch

Related Services

obs

Live Subtitles

Live Subtitles

ASR can convert the audio from a live video stream into audience-friendly subtitles in real time and conduct quality checks based on the Text Moderation service.

Advantages

High Speed

Real-time live speech recognition

High Accuracy

High recognition accuracy

Related Services

obs

Human-Machine Interaction

Human-Machine Interaction

ASR integrates the voice wakeup service so that voice commands sent to terminals initiate operations in real time, improving the interaction experience between human and machine.

Advantages

High Wakeup Rate

High device activation rate by human voice commands

Customization

Allows customized wakeup keywords

Related Services

obs

Functions

  • Speech Recognition

    Supports the recognition of short-sentence, long-sentence, and farfield speech segments and converts speech into text in real time.

  • Language Support

    Supports Mandarin Chinese recognition.

Speech Recognition

Language Support

  • Voice Wakeup

    Devices in a locked or hibernating state can be woken by voice commands.

  • Voice Command Capability

    Sends operation commands to devices by voice.

Voice Wakeup

Voice Command Capability

Usage Guides
Documentation

Technical Documents

Automatic Speech Recognition User Guide

FAQs

How Can I Use ASR APIs?
Huawei Teams Up with iFLYTEK to Offer Improved Voice Service Experience

Create an Account and Experience HUAWEI CLOUD for Free

Register Now