Verbit Software Ltd.

Alcatel Lucent

Alcatel-lucent USA Inc.

Elisha Yehuda Rosensweig

Tom Livne

Eric Ariel Shellef

Yaakov Kobi Ben Tsvi

Iris Getz

Randeep Singh Bhatia

Bhawna Gupta

Etti Shalev

Roman Himmelreich

Eli Asor

Dmitri Krasnenko

Ifat Afek

Sharon Mendel-Brin

When transcribing audio recordings, such as legal depositions, phrases may be repeated throughout the recordings, but these repeated phrases get transcribed incorrectly by an automatic speech recognition (ASR) system. In order to assist a transcriber to correctly resolve such phrases, some embodiments described herein involve a computer that receives an audio recording that includes speech, generates a transcription of the audio recording utilizing an ASR system, and clusters segments of the audio recording into clusters of similar utterances. The computer provides a transcriber with certain segments of the audio recording, which include similar utterances belonging to a certain cluster, along with transcriptions of the certain segments. The computer receives from the transcriber: an indication of which of the certain segments include repetitions of a phrase, and a correct transcription of the phrase. The computer then updates the transcription of the audio recording based on the indication and the correct transcription.

Human resolution of repeated phrases in a hybrid transcription system

Maintaining adequate audio quality is very important for creating fast and accurate transcriptions, especially in a hybrid transcription setting, in which human transcribers review transcriptions generated by automatic speech recognition (ASR) systems. Some embodiments described herein involve detecting low-quality audio intended for transcription. In one embodiment, a server receives an audio recording that includes speech. The server generates feature values based on a segment of the audio recording and utilizes a model to calculate, based on the feature values, a certain value indicative of expected hybrid transcription quality of the segment. The model is generated based on training data that includes feature values generated based on previously recorded segments of audio, and values of transcription-quality metrics generated based on transcriptions of the previously recorded segments, which were generated at least in part by human transcribers. Optionally, an alert is provided responsive to the certain value being below a threshold.

Real time machine learning-based indication of whether audio quality is suitable for transcription

Hybrid transcription of audio relies on having one or more layers of transcribers who review transcriptions generated by automatic speech recognition (ASR) systems in order to correct errors that are found in the transcriptions. When it comes to determining how much human reviewing is needed, such as determining how many layers of review to use, there is a cost/benefit tradeoff to consider. Some embodiments described herein utilize a machine learning-based approach for estimating quality of hybrid transcription of audio. In one embodiment, a computer generates a transcription of a segment of audio using an ASR system, which is subsequently reviewed by a transcriber. The computer then calculates, based on properties of the review by the transcriber, a value indicative of an expected accuracy of the reviewed transcription. The computer may suggest a second transcriber review the reviewed transcription if the value indicative of the expected accuracy is below a threshold.

Quality estimation of hybrid transcription of audio

When transcribing large audio files, such as in the case of legal depositions, there are often many transcribers to choose from. Embodiments described herein enable calculation of expected accuracy of transcriptions by transcribers, which can be used to guide the selection of transcribers for specific tasks. In one embodiment, a computer receives a segment of an audio recording that includes speech of a person, and identifies an accent of the person and a topic of the segment. The computer generates feature values based on data that includes the accent and the topic, and utilizes a model to calculate, based on the feature values, an expected accuracy of a transcription of the segment by a certain transcriber. The model is generated based on training data that includes segments of previous audio recordings and values of accuracies of transcriptions, by the certain transcriber, of the segments.

Machine learning-based prediction of transcriber performance on a segment of audio

Various embodiments provide a method and apparatus for allocating resources to applications (e.g., application processes) using pre-provisioned resource configurations (e.g., VMs or containers) that are based on clustered historical resource usage data. In particular, the set of pre-provisioned resources required to support elasticity in a set of applications is reduced by determining a subset of pre-provisioned resource configurations that are able to be reused across multiple applications. A pre-provisioned resource configuration being based on a determined dominant center for the resource cluster associated with the pre-provisioned resource configuration.

Method and apparatus for provisioning resources using clustering

A method and apparatus for substantially minimizing overhead over-provisioning costs in machine configurations is disclosed. The method involves the steps of: clustering a plurality of given machine configurations into a quantity of clusters less than or equal to a pre-specified amount; determining a respective dominant provisioning machine configuration for each cluster of the quantity of clusters; and determining an overall over-provisioning resource cost associated with the respective quantity of clusters and associated respective dominant provisioning machine configurations; and assigning to a mapping function the respective associated dominant provisioning machine configuration of each respective cluster of the quantity of clusters as the target for the given machine configurations of each respective cluster of the quantity of clusters. The method for substantially minimizing overhead over-provisioning costs in machine configurations provides advantages over systems known in the art by allowing minimization of average overhead due to over-provisioning costs as well as minimizing the maximum overhead due to over-provisioning costs.

Growing community of inventors

Raanana, Israel

Elisha Yehuda Rosensweig