Voice-Controlled Device Speech Output Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Voice-controlled devices face challenges in determining when to initiate or stop conversations due to uncertainty about whether an utterance is directed at them and whether it is appropriate to speak in the presence of humans or other devices, leading to poor interaction with humans and other voice-controlled devices.
Innovation Solution
The system uses sensors to detect the number of people within a settable distance and determines whether to output synthesized speech based on audio input, requiring an attention word only when multiple people are present, and pauses speech if ambient noise exceeds a certain level, resuming once noise decreases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the voice-controlled device waits for an attention word before outputting synthesized speech, then it avoids inappropriate interruptions, but it increases response time and may miss conversational opportunities
Solution Approach 1:
The system dynamically adjusts its response behavior based on the detected social context. When multiple people are detected, it requires an attention word before responding. When a single person is detected, it can respond more immediately without waiting for an attention word, thus adapting its timing strategy to the social situation.
Solution Approach 2:
The system changes the parameter of response threshold based on the number of detected people. In multi-person environments, it maintains a higher threshold (requiring attention word). In one-person environments, it lowers the threshold (responding without attention word), optimizing both appropriateness and response time.
2Adaptability or versatility
If the voice-controlled device speaks in the presence of multiple people, then it responds to all potential users, but it may cause confusion about who is being addressed
Solution Approach 1:
The attention word serves as an intermediary signal that resolves ambiguity in multi-person environments. When multiple people are present, the system requires an attention word to be spoken before responding, which acts as a mediator to indicate which person's input should be addressed.
Solution Approach 2:
The system segments its response behavior based on the number of detected people. In multi-person scenarios, it segments the communication process by requiring an attention word to identify the intended speaker, thereby separating the detection phase from the response phase.
3Reliability
If the voice-controlled device continuously monitors audio input to detect when to pause speech, then it can respond to interruptions appropriately, but it increases processing load and complexity
Solution Approach 1:
The system implements a feedback mechanism where it continuously monitors audio input during speech output and uses this feedback to determine when to pause. When audio input exceeding a threshold is detected, the system pauses its synthesized speech, creating a closed-loop control system that adapts to real-time acoustic conditions.
4Ease of operation
If the voice-controlled device apologizes for interruptions, then it improves social interaction naturalness, but it increases the length of interaction and may disrupt conversation flow
Solution Approach 1:
The system performs a preliminary anti-action by apologizing before resuming speech after an interruption. This preemptive apology acknowledges the disruption and maintains social etiquette, making the interaction more natural while minimizing further disruption by quickly returning to the original speech flow.
Data Source
AI summary
A system, a computer program product, and method for controlling synthesized speech output on a voice-controlled device. One or more sensors are used to detect whether one person or more than one person is within a first settable distance from the voice-controlled device. Next a determination is made whether the audio input received is recognized as speech. In response to only one person being detected within the settable distance, begin outputting synthesized speech based on the audio input without waiting for an attention word to be recognized and otherwise wait for additional criteria before outputting synthesized speech based on the speech input. The additional criteria includes determining that more than one person is detected and recognizing that the attention word is received before outputting synthesized speech based on the audio input. Another criteria includes more than one person being detected and determining that the person is looking at the voice-controlled device before outputting synthesized speech based on the audio input.