Speech Encoder Adaptive Band-Limiting for Low Bit Rate Quality

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In speech encoding systems, there is a trade-off between the sound quality of wide band and extended band signals due to bit rate constraints, leading to overall sound quality degradation when encoding super-wide band signals at low bit rates.

Innovation Solution

A speech encoding and decoding apparatus that adaptively selects and limits frequency bands in the extended band layer based on the pitch period of the input signal, reducing quantization noise by attenuating energy in higher frequency bands, thereby maintaining sound quality even at low bit rates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a relatively high bit rate is allocated for encoding the wide band signal, then the sound quality of the decoded wide band signal is improved, but the bit rate allocated for encoding the extended band signal becomes very low, resulting in quantization noise and overall sound quality degradation

Engineering Contradiction:
Improvesound quality of decoded wide band signalVSAvoidquantization noise in extended band signal
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent dynamically changes the band-limiting frequency parameter based on the pitch period of the input signal. When the pitch period indicates a voiced sound, the band-limiting frequency is lowered to reduce quantization noise in the extended band. When the pitch period indicates an unvoiced sound, the band-limiting frequency is raised to preserve extended band quality. This dynamic parameter adjustment resolves the contradiction by adapting the encoding strategy to the actual signal characteristics.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces dynamic adaptation of the band-limiting frequency based on pitch period detection. The encoding system transitions from a static band-limiting approach to a dynamic one where the limiting frequency changes according to the signal's pitch characteristics. This allows the system to optimize the trade-off between wide band and extended band quality in real-time, preventing quantization noise in voiced regions while maintaining quality in unvoiced regions.

Inventive Principle:
Principle #15Dynamics

2Loss of information

If the bit rate used for encoding of the wide band signal is decreased, then the bit rate allocated for encoding of the extended band signal is increased, but the sound quality of decoded wide band signal is degraded, resulting in overall sound quality degradation

Engineering Contradiction:
Improvequantization noise in extended band signalVSAvoidsound quality of decoded wide band signal
Core Design Contradiction:
Loss of informationVSMeasurement precision

Solution Approach 1:

The patent uses parameter changes in the band-limiting frequency to control the distribution of encoding resources. By lowering the band-limiting frequency during voiced periods, the system effectively reduces the encoding burden on the extended band while maintaining adequate wide band quality. This dynamic parameter adjustment allows flexible bit rate allocation that prevents both types of quality degradation.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If encoding of speech signal including a super-wide band signal is done at a low bit rate, then there is a trade-off between the sound quality of the decoded wide band signal and the sound quality of the decoded extended band signal

Engineering Contradiction:
Improvebit rate efficiencyVSAvoidoverall sound quality
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent implements dynamic band-limiting frequency adjustment based on pitch period detection to optimize low bit rate encoding. The system adaptively changes the encoding strategy according to whether the signal is voiced or unvoiced, maximizing sound quality under bit rate constraints. This dynamic approach eliminates the need for a fixed trade-off by optimizing the encoding parameters in real-time according to signal characteristics.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS9264094B2Voice coding device, voice decoding device, voice coding method and voice decoding method
Publication Date: 2016.02.16 PANASONIC INTELLECTUAL PROPERTY CORP OF AMERICA
  • US9264094B2 patent drawing
  • US9264094B2 patent drawing
  • US9264094B2 patent drawing

AI summary

A voice coding device capable of preventing overall quality degradation even when the bit rate for coding is lowered. The voice coding device codes a wide band signal in a first layer, and codes an extended band signal whose frequency band is located in higher frequency than the wide band signal in an extended band layer. An adaptive band selection unit (301) selects a frequency band to be excluded from a coding object in the extended band layer or a frequency band whose energy is to be attenuated in the extended band layer. A band-limited signal generation unit (302) excludes, within the frequency band of an input signal, the frequency band selected by the adaptive band selection unit (301) from the coding object, or attenuates the energy of the frequency band selected by the adaptive band selection unit (301).