Speech Compression by Phoneme Segmentation and Pitch Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech signal compression methods, such as entropy-coding, face low compression efficiency due to the lack of clear regularity in speech waveforms and pitch fluctuations, making it difficult to compress speech data effectively, especially in devices with limited storage capacity.
Innovation Solution
A speech signal compression device that divides speech signals into phoneme portions, extracts and adjusts pitch signals, generates sub-band data, and performs data compression based on predetermined conditions for each phoneme, using nonlinear quantization and prioritizing spectral components to achieve efficient compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If entropy-coding method is used to compress speech data, then data capacity is reduced, but compression efficiency remains low due to lack of clear regularity in speech waveforms
Solution Approach 1:
The speech signal is divided into multiple sections based on pitch period boundaries. Each section corresponds to one pitch cycle and is processed independently for entropy coding. This segmentation reveals regularity within each pitch period while accommodating pitch fluctuations between periods, thereby improving compression efficiency while reducing data capacity.
2Ease of operation
If speech data is separated at regular intervals of time length, then processing is simplified, but separation timing does not correspond to phoneme boundaries resulting in low compression efficiency
Solution Approach 1:
The system uses pitch detection feedback to dynamically determine section boundaries. By detecting pitch periods and using this information to guide the segmentation of speech data into sections, the method aligns processing boundaries with acoustic features (phoneme boundaries) rather than using fixed regular intervals. This feedback mechanism maintains processing simplicity while significantly improving compression efficiency.
3Device complexity
If pitch fluctuation is not considered, then processing is simpler, but compression efficiency decreases because pitch length is not constant
Solution Approach 1:
The system dynamically adjusts section boundaries based on detected pitch periods rather than using fixed intervals. Each section is defined by consecutive pitch periods, allowing the segmentation to adapt to pitch fluctuations caused by emotion or consciousness. This dynamic approach handles variable pitch lengths while maintaining processing simplicity through automated pitch detection.
Data Source
AI summary
The present invention provides a speech signal compression device which allows a storage capacity of data representing speech to be efficiently compressed. In the present invention, a computer C1 operates with respect to speech data to be compressed into speech data for each phoneme on the basis of phoneme labeling data, to unify the time length of a unit pitch section for each of the divided speech data into the same value, thereby creating a pitch waveform and creating a sub-band data representing variation in time of spectrum components of the pitch waveform signal. Also, this sub-band data is compressed so as to match a condition designated by a table for compression, and the compressed data is further encoded in entropy to output the entropy coded data.