- Release Notes
- Product Introduction
- User Tutorial
- Purchase Guide
- SDK Download Guide
- Product Feature Demo
- Console Guide
- Getting Started
- Basic Feature Development Guide
- Advanced Feature Development Guide
- Client API
- Server APIs
- FAQ
- Service Agreement
- Contact Us
- Glossary
- GME Policy
- Release Notes
- Product Introduction
- User Tutorial
- Purchase Guide
- SDK Download Guide
- Product Feature Demo
- Console Guide
- Getting Started
- Basic Feature Development Guide
- Advanced Feature Development Guide
- Client API
- Server APIs
- FAQ
- Service Agreement
- Contact Us
- Glossary
- GME Policy
Voice Tag
Last updated: 2024-01-18 11:35:26
GME's voice tag service analyzes the input audio data based on AI models to add tags such as age, gender, and tone to the audio.
Overview
Age tag: GME recognizes the age range of players.
Gender tag: GME recognizes the gender of players.
Tone tag: GME recognizes the tone of players.
Use cases
1. Gaming feature rating
GME allows you to rate gaming features by age; for example, you can provide the minor mode and adult mode in your game.
GME enables rated management of ad content and game stories.
2. Minor protection
GME uses voice recognition to detect minor players in team battling, so as to find minors bypassing identity verification and the anti-addiction system.
For accounts publishing non-compliant content involving minors, such as sensitive content, privacy, sexually suggestive content, and violent content, GME quickly locates and blocks them to prevent the risky information involving minors from being spread.
3. User characteristic analysis
GME summarizes users' characteristics according to their age, gender, and tone and analyzes such characteristics to facilitate game operations.
Product strengths
1. High accuracy and recall rate
GME delivers industry-leading accuracy and recall rates. In game voice scenarios, the accuracy and recall rate reach at least 95%. It guarantees the recognition effect by extracting the voiceprint characteristics of the audio signals from massive amounts of real scenario data and leveraging the deep learning system based on long short-term memory (LSTM). Plus, the AI model is optimized in depth for various voice chat-based gaming scenarios, such as Werewolf, team battling, and game audio interaction, so as to effectively recognize their tags.
2. High performance
GME supports a high number of concurrent recognition requests and requests for large files. It can be scaled horizontally and deployed in multiple regions to guarantee that recognition tasks are not blocked.
3. Support for various languages
GME's voice tag recognition model supports various languages worldwide.
4. Security compliance and privacy protection
GME does not require user identity information to assist with recognition or retain moderated audio to protect the user privacy.
Yes
No
Was this page helpful?