Speech Signal Alignment Using Segmented Phone Sequence Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The alignment of speech data with transcripts is challenging, especially for long speech data, as existing methods struggle with accuracy and segmentation, particularly when transcripts are modified or not segmented appropriately for effective alignment.
Innovation Solution
A computer-implemented method that segments both speech signals and transcripts into utterance-like units, performs automatic speech recognition, converts them into phone sequences, calculates correspondence costs, and generates mapping data to determine the optimal alignment series, thereby improving the accuracy of speech data alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If speech data is segmented into utterances by referring to pauses, then alignment accuracy is improved, but the process becomes complex and time-consuming
Solution Approach 1:
The patent applies segmentation by dividing the speech data into multiple utterances based on detected pauses, and further dividing transcripts into phrase units. This creates manageable segments that can be aligned independently, improving alignment accuracy while making the overall process more systematic and less complex through automated segmentation rules.
2Measurement precision
If speech data is segmented into utterances by referring to pauses, then alignment accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary segmentation of both speech data and transcripts into manageable units (utterances and phrase units) before alignment. By preparing these segments in advance with automated pause detection and phrase division, the actual alignment process becomes faster and more efficient, reducing overall processing time while maintaining high accuracy.
3Ease of operation
If straightforward aligning method is used on unsegmented transcripts, then processing is simpler, but alignment accuracy deteriorates
Solution Approach 1:
The patent segments transcripts into phrase units that correspond to utterances in the speech data. This segmentation enables a systematic alignment process that maintains simplicity while improving accuracy, as the aligned units are smaller and more manageable, allowing for more precise correspondence between speech and text.
4Ease of operation
If transcripts are modified or deleted for better readability, then readability is improved, but alignment capability deteriorates
Solution Approach 1:
The patent performs partial alignment by dividing transcripts into phrase units and aligning only the necessary portions with corresponding utterances. This partial action approach maintains readability modifications while preserving alignment capability for the segments that need to be aligned, rather than requiring complete transcript fidelity.
Data Source
AI summary
A method, a computer program product, and a computer system for mapping between a speech signal and a transcript of the speech signal. The computer system segments the speech signal to obtain one or more segmented speech signals and the transcript of the speech signal to obtain one or more segmented transcripts of the speech signal. The computer system generates estimated phone sequences and reference phone sequences, calculates costs of correspondences between the estimated phone sequences and the reference phone sequences, determines a series of the estimated phone sequences with a smallest cost, selects a partial series of the estimated phone sequences from the series of the estimated phone sequences, and generates mapping data which includes the partial series of the estimated phone sequences and a corresponding series of the reference phone sequences.