Parallel Uplink Downlink Speech Recognition Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition technologies using multiple HTTP connections struggle with real-time data upload and download due to the need for data reintegration and high traffic consumption.
Innovation Solution
Implementing a speech recognition method and device that utilize parallel uplink and downlink connections for streaming data transmission, allowing real-time upload and download of speech data and recognition results via separate HTTP connections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If speech data is sent via multiple HTTP connections to a speech server for reintegration and processing, then speech recognition can be performed, but real-time upload and down sending of data cannot be realized and network traffic consumption increases
Solution Approach 1:
The patent divides the data transmission process into separate uplink and downlink connections. The uplink connection is dedicated for sending speech data to the server, while the downlink connection is dedicated for receiving recognition results back to the client. This segmentation allows simultaneous bidirectional communication without data reintegration, achieving real-time transmission while reducing network traffic by eliminating redundant HTTP headers and reintegration overhead.
2Productivity
If speech data is sent via multiple HTTP connections for processing, then speech recognition can be performed, but data reintegration is required which prevents real-time transmission
Solution Approach 1:
The patent establishes separate uplink and downlink connections before data transmission begins. The server prepares to receive speech data continuously on the uplink connection while simultaneously preparing to send recognition results on the downlink connection. This preliminary setup eliminates the need for data reintegration during transmission, allowing real-time processing and transmission to proceed in parallel without time loss.
3Adaptability or versatility
If traditional HTTP connections are used for speech recognition, then compatibility is maintained, but structure consumption such as HTTP head increases traffic flow
Solution Approach 1:
The patent utilizes the existing HTTP protocol framework while introducing a dual-connection architecture that serves multiple functions simultaneously. Both uplink and downlink connections use standard HTTP protocols, maintaining compatibility with existing systems. However, the parallel connection structure enables efficient streaming data transmission and result delivery without requiring additional specialized protocols, thus reducing traffic flow by eliminating redundant HTTP headers while preserving protocol universality.
Data Source
AI summary
The present disclosure provides a speech recognition method, device and system based on artificial intelligence. The method includes: collecting speech data to be recognized in a speech recognition process; sending uplink data stream to a server via an uplink connection to the server, in which the uplink data stream includes the speech data; and receiving downlink data stream sent by the server via a downlink connection to the server in parallel with sending the uplink data stream to the server, in which the downlink data stream includes result data, and the result data is obtained by the server performing speech recognition according to the speech data.