Impulse Noise Correction in Speech Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
First responders face delays and increased processing demands due to impulse noise interfering with speech recognition systems, causing incomplete utterance recognition and repeated requests in noisy environments, which can be critical in emergency situations.
Innovation Solution
The system detects impulse noise in utterances and generates clarifying voice prompts to confirm or repeat affected words, reducing the need for users to repeat entire utterances, thereby improving interaction and resource efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If speech recognition systems process utterances in noisy environments, then the system can provide information and services to first responders, but impulse noise causes incomplete utterance recognition and requires repeated requests, increasing processing demands and delays
Solution Approach 1:
The system performs preliminary action by detecting impulse noise and generating clarifying voice prompts before the user needs to repeat their entire utterance. The electronic processor identifies portions of utterances affected by impulse noise and proactively requests clarification on specific words or phrases, preventing incomplete recognition from propagating through the system and reducing subsequent reprocessing needs.
2Reliability
If the system requests users to repeat entire utterances due to impulse noise, then recognition accuracy may be maintained, but communication efficiency decreases and battery resources are consumed
Solution Approach 1:
The system applies segmentation by dividing the utterance into discrete portions and identifying specific segments affected by impulse noise. Instead of requiring repetition of the entire utterance, the electronic processor generates clarifying voice prompts that target only the noisy segments or specific words/phrases that were missed, reducing the amount of audio data that needs to be reprocessed and thereby conserving battery resources.
3Productivity
If the system processes every utterance completely despite impulse noise, then no clarification requests are needed, but critical information may be missed and processing resources are wasted
Solution Approach 1:
The system implements feedback by monitoring the output of speech recognition and detecting when impulse noise has degraded utterance quality. The electronic processor analyzes recognition confidence and audio quality metrics, then provides feedback in the form of targeted clarifying voice prompts for specific portions of the utterance. This closed-loop approach ensures critical information is captured while avoiding unnecessary reprocessing of clear segments.
Data Source
AI summary
System and method for correcting for impulse noise in speech recognition systems. One example system includes a microphone, a speaker, and an electronic processor. The electronic processor is configured to receive an audio signal representing an utterance. The electronic processor is configured to detect, within the utterance, the impulse noise, and, in response, generate an annotated utterance including a timing of the impulse noise. The electronic processor is configured to segment the annotated utterance into silence, voice content, and other content, and, when a length of the other content is greater than or equal to an average word length for the annotated utterance, determine, based on the voice content, an intent portion and an entity portion. The electronic processor is configured to generate a voice prompt based on the timing of the impulse noise and the intent portion and/or the entity portion, and to play the voice prompt.