|
Work item:
|
H.MVIS
|
|
Subject/title:
|
Functional architecture of multimodal enabled vehicular voice interaction system
|
|
Status:
|
Under study
|
|
Approval process:
|
AAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2025 (Medium priority)
|
|
Liaison:
|
-
|
|
Supporting members:
|
-
|
|
Summary:
|
After the IMT-2020 network technology spread, it is expected that voice interactive control technology is widely used in the field of intelligent cockpit. This technology provides great convenience for drivers and passengers when they're interacting with the vehicle.
However, there are still many users' pain points in the field of single mode speech interaction recognition. For example, accurate recognition rate is relatively low, false recognition and high latency. In order to solve the above pain points, the industry is developing AI-based multi-modal joint recognition schemes.
The purpose of this work item is to figure out the reference model and core architecture of the H.MVIS with AI-based multi-modal joint recognition functions to improve human speech recognition rate in cockpit and to define the requirements of this H.MVIS.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2023-08-21 10:14:30
|
|
Last update:
2024-06-11 10:19:48
|
