DTMF Signal Erasure in Speech Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current DTMF signal processing methods in bidirectional communication systems introduce significant delay due to the need to decode and accumulate speech data to identify and erase DTMF signals, which impedes serviceability and can be further complicated by packet loss or corruption in networks.
Innovation Solution
A DTMF signal processing apparatus that divides speech data into units, analyzes for DTMF components, applies weighting values, and determines whether to replace the data with mute or noise, allowing for real-time erasure without accumulating speech data, thus minimizing delay.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If speech data is decoded and accumulated for a certain section to determine whether it contains a DTMF signal, then the DTMF signal can be accurately detected, but a significant delay in speech is inevitable
Solution Approach 1:
The patent segments speech data into individual packets and processes them sequentially without accumulating multiple sections. By analyzing each packet independently as it arrives, the system achieves DTMF detection without the time delay caused by traditional methods that require accumulating a certain section of speech data.
Solution Approach 2:
The patent applies weighting to spectral components in advance before DTMF detection. By pre-calculating and storing weighted spectral data for each speech data packet, the system prepares the necessary information beforehand, enabling rapid DTMF signal determination without requiring subsequent accumulation or reprocessing of multiple data sections.
2Object-generated harmful factors
If the DTMF signal is erased by replacing speech data with mute or noise data, then the DTMF signal is removed from the communication stream, but the processing complexity increases
Solution Approach 1:
The patent extracts and identifies DTMF signals from the speech data stream by analyzing spectral components, then removes only the identified DTMF portions by replacing them with mute or noise data. This selective extraction approach eliminates the harmful DTMF signal while preserving the rest of the speech data, avoiding the need for complex full-signal processing.
Solution Approach 2:
The patent changes the state of speech data packets by applying weighting factors to spectral components and using these weighted values to detect and erase DTMF signals. By transforming the detection criterion from time-domain analysis to frequency-domain weighted analysis, the system simplifies the erasure process while improving detection accuracy.
3Loss of time
If speech data is processed in real-time without accumulation, then speech delay is minimized, but the reliability of DTMF signal detection may be affected by packet loss or corruption
Solution Approach 1:
The patent prepares weighted spectral components in advance for each speech data packet and stores them before transmission or further processing. This preliminary preparation creates a buffer of pre-processed data that can be reliably analyzed even if packet loss or corruption occurs during transmission, maintaining detection reliability without requiring post-receipt accumulation of multiple packets.
Data Source
AI summary
A DTMF signal processing apparatus of the present invention comprises a data divider unit, a DTMF signal component analyzer unit, a weighting processing unit, a buffer, a DTMF signal erasure determination unit, and a DTMF signal erasure processing unit. The data divider unit divides speech data into a plurality of divided speech data, and the DTMF signal component analyzer unit analyzes whether or not the divided speech data has a DTMF signal component. The weighting processing unit applies a weighting value to divided speech data analyzed at this time and stores the resultant speech data in the buffer, and also applies a weighting value to past divided speech data previously stored in the buffer when the result of the analysis indicates that the analyzed divided speech data has the DTMF signal component. The DTMF signal erasure determination unit determines based on the weighting value whether or not to erase the divided speech data stored in the buffer. The DTMF signal erasure processing unit replaces the divided speech data with either mute data or noise data, and delivers the replaced data when the result of the determination indicates an erasure.