Google Inc.

Yiling Huang

Quan Wang

Jason William Pelecanos

Pu-sen Chao

Ignacio Lopez Moreno

Alanna Foster Slocum

Mert Saglam

Shelly Bensal

Pu-Sen Chao

A method includes obtaining a speaker identification (SID) model trained to predict speaker embeddings from utterances spoken by different speakers, the SID model includes a trained audio encoder and a trained SID head. The method also includes receiving a plurality of synthetic speech detection (SSD) training utterances that include a set of human-originated speech samples and a set of synthetic speech samples. The method also includes training, using the trained audio encoder, a SSD head on the SSD training utterances to learn to detect the presence of synthetic speech in audio encodings encoded by the trained audio encoder. The operations also include providing, for execution on a computing device, a multitask neural network model for performing both SID tasks and SSD tasks on input audio data in parallel.

Speaker verification with multitask speech models

A method for evaluating a verification model includes receiving a first and a second set of verification results where each verification result indicates whether a primary model or an alternative model verifies an identity of a user as a registered user. The method further includes identifying each verification result in the first and second sets that includes a performance metric. The method also includes determining a first score of the primary model based on a number of the verification results identified in the first set that includes the performance metric and determining a second score of the alternative model based on a number of the verification results identified in the second set that includes the performance metric. The method further includes determining whether a verification capability of the alternative model is better than a verification capability of the primary model based on the first score and the second score.

Assessing speaker recognition performance

A speaker verification method includes receiving audio data corresponding to an utterance, processing the audio data to generate a reference attentive d-vector representing voice characteristics of the utterance, the evaluation ad-vector includes nstyle classes each including a respective value vector concatenated with a corresponding routing vector. The method also includes generating using a self-attention mechanism, at least one multi-condition attention score that indicates a likelihood that the evaluation ad-vector matches a respective reference ad-vector associated with a respective user. The method also includes identifying the speaker of the utterance as the respective user associated with the respective reference ad-vector based on the multi-condition attention score.

Attentive scoring function for speaker identification

Growing community of inventors

Mountain View, CA, United States of America

Yiling Huang