← Back to papers
2022 INTERSPEECH INTERSPEECH 2022

The ZevoMOS entry to VoiceMOS Challenge 2022

Abstract

This paper introduces the ZevoMOS entry to the main track of the VoiceMOS Challenge 2022. The ZevoMOS submission is based on a two-step finetuning of pretrained self-supervised learning (SSL) speech models. The first step uses a task of classifying natural versus synthetic speech, while the second step's task is to predict the MOS scores associated with each training sample. The results of the finetuning process are then combined with the confidence scores extracted from an automatic speech recognition model, as well as the raw embeddings of the training samples obtained from a wav2vec SSL speech model. The team id assigned to the ZevoMOS system within the VoiceMOS Challenge is T01. The submission was placed on the 14th place with respect to the system-level SRCC, and on the 9th place with respect to the utterance-level MSE. The paper also introduces additional evaluations of the intermediate results.

🌉 Interdisciplinary Bridge — Deep Learning and Machine Learning and Speech & Audio
🧭 Keyword Pioneer — mean opinion score prediction
🐝 Cross-Pollinator — Artificial Intelligence, Computer Science, Computer Vision, Data Science & Analytics, Deep Learning, Healthcare & Medicine, Interdisciplinary, Knowledge & Reasoning, Machine Learning, Mathematics & Optimization, Natural Language Processing, Reinforcement Learning, Robotics, Security & Privacy, Speech & Audio

Authors

Adriana Stan

Topics

Machine Learning > Learning Types > Self-Supervised Learning Deep Learning > Techniques > Pretraining Speech & Audio > Analysis > Prosody Analysis

Keywords

transfer learning self-supervised learning pretrained model speech quality assessment speech model mos prediction mean opinion score prediction
Download PDF

Related papers

Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis 2022
Which Model is Best: Comparing Methods and Metrics for Automatic Laughter Detection in a Naturalistic Conversational Dataset 2022
Evidence of Onset and Sustained Neural Responses to Isolated Phonemes from Intracranial Recordings in a Voice-based Cursor Control Task 2022
Pre-trained Speech Representations as Feature Extractors for Speech Quality Assessment in Online Conferencing Applications 2022
Exploring the influence of fine-tuning data on wav2vec 2.0 model for blind speech quality prediction 2022