Back to results
US 10,433,052 B2Granted
System and method for identifying speech prosody
OwnerRon Zass, Yotam Zass Rozenfeldinventors
Life of this patent
Estimating the term…
Abstract
System and method for analyzing audio data are provided. The audio data may be analyzed to identify speech prosody. For example, the audio data may be analyzed to select a portion of the audio data containing speech produced by a first speaker. The audio data may be further analyzed to identify speech prosody of the speech within the selected portion. Feedbacks and reports may be provided based on the identified speech prosody.
Claims25
A system for processing audio for identifying loudness levels of speakers, the system comprising:
- one or more audio sensors included in a wearable apparatus and configured to capture audio data from an environment of a wearer of the wearable apparatus;
- and
- at least one processing unit configured to:
- analyze the audio data to determine that the wearer and a second speaker are engaged in conversation;
- analyze the audio data to obtain prosodic information associated with the wearer and at least part of the conversation, the prosodic information comprises an indication of a loudness level;
- estimate a distance between the wearer and the second speaker;
- use the estimated distance between the wearer and the second speaker and the indication of the loudness level to determine that the loudness level is inappropriate for the estimated distance;
- in response to the determination that the loudness level is inappropriate for the estimated distance and the determination that the wearer and the second speaker are engaged in conversation, provide a feedback to the wearer;
- obtain additional audio data captured by the one or more audio sensors after obtaining the prosodic information, the additional audio data comprising an additional part of the conversation;
- estimate a second distance between the wearer and the second speaker;
- analyze the additional audio data to obtain additional prosodic information associated with the wearer, the additional prosodic information comprises an additional indication of an additional loudness level;
- use the estimated second distance between the wearer and the second speaker and the additional indication of the additional loudness level to determine that the additional loudness level is inappropriate for the estimated second distance;
- determine an elapsed time since the feedback was provided to the wearer;
- when the elapsed time is longer than a selected time duration, in response to the determination that the additional loudness level is inappropriate for the estimated second distance, provide an additional feedback to the wearer;
- and
- when the elapsed time is shorter than the selected time duration forgo the additional feedback to the wearer.
13 more elements
A method for processing audio for identifying loudness levels of speakers, the method comprising:
- capturing audio data using one or more audio sensors included in a wearable apparatus from an environment of a wearer of the wearable apparatus;
- analyzing the audio data to determine that the wearer and a second speaker are engaged in conversation;
- analyzing the audio data to obtain prosodic information associated with the wearer and at least part of the conversation, the prosodic information comprises an indication of a loudness level;
- estimating a distance between the wearer and the second speaker;
- using the estimated distance between the wearer and the second speaker and the indication of the loudness level to determine that the loudness level is inappropriate for the estimated distance;
- in response to the determination that the loudness level is inappropriate for the estimated distance and the determination that the wearer and the second speaker are engaged in conversation, providing a feedback to the wearer;
- obtaining additional audio data captured by the one or more audio sensors after obtaining the prosodic information, the additional audio data comprising an additional part of the conversation;
- estimating a second distance between the wearer and the second speaker;
- analyzing the additional audio data to obtain additional prosodic information associated with the wearer, the additional prosodic information comprises an additional indication of an additional loudness level;
- using the estimated second distance between the wearer and the second speaker and the additional indication of the additional loudness level to determine that the additional loudness level is inappropriate for the estimated second distance;
- determining an elapsed time since the feedback was provided to the wearer;
- when the elapsed time is longer than a selected time duration, in response to the determination that the additional loudness level is inappropriate for the estimated second distance, providing an additional feedback to the wearer;
- and
- when the elapsed time is shorter than the selected time duration, forgoing the additional feedback to the wearer.
11 more elements
A non-transitory computer readable medium storing data and computer implementable instructions for carrying out a method for processing audio for identifying loudness levels of speakers, the method comprising:
- capturing audio data using one or more audio sensors included in a wearable apparatus from an environment of a wearer of the wearable apparatus;
- analyzing the audio data to determine that the wearer and a second speaker are engaged in conversation;
- analyzing the audio data to obtain prosodic information associated with the wearer and at least part of the conversation, the prosodic information comprises an indication of a loudness level;
- estimating a distance between the wearer and the second speaker;
- using the estimated distance between the wearer and the second speaker and the indication of the loudness level to determine that the loudness level is inappropriate for the estimated distance;
- in response to the determination that the loudness level is inappropriate for the estimated distance and the determination that the wearer and the second speaker are engaged in conversation, providing a feedback to the wearer;
- obtaining additional audio data captured by the one or more audio sensors after obtaining the prosodic information, the additional audio data comprising an additional part of the conversation;
- estimating a second distance between the wearer and the second speaker;
- analyzing the additional audio data to obtain additional prosodic information associated with the wearer, the additional prosodic information comprises an additional indication of an additional loudness level;
- using the estimated second distance between the wearer and the second speaker and the additional indication of the additional loudness level to determine that the additional loudness level is inappropriate for the estimated second distance;
- determining an elapsed time since the feedback was provided to the wearer;
- when the elapsed time is longer than a selected time duration in response to the determination that the additional loudness level is inappropriate for the estimated second distance, providing an additional feedback to the wearer;
- and
- when the elapsed time is shorter than the selected time duration, forgoing the additional feedback to the wearer.
11 more elements
Dependencies are inferred from the claim text.