{"id":682,"date":"2023-08-21T09:14:34","date_gmt":"2023-08-21T09:14:34","guid":{"rendered":"https:\/\/todaysainews.com\/index.php\/2023\/08\/21\/google-at-interspeech-2023-google-research-blog\/"},"modified":"2025-04-27T07:33:00","modified_gmt":"2025-04-27T07:33:00","slug":"google-at-interspeech-2023-google-research-blog","status":"publish","type":"post","link":"https:\/\/todaysainews.com\/index.php\/2023\/08\/21\/google-at-interspeech-2023-google-research-blog\/","title":{"rendered":"Google at Interspeech 2023 \u2013 Google Research Blog"},"content":{"rendered":"

[ad_1]
\n<\/p>\n

\nPosted by Catherine Armato, Program Manager, Google<\/span><\/p>\n

<\/p>\n

\nThis week, the 24th Annual Conference of the International Speech Communication Association<\/a> (INTERSPEECH 2023) is being held in Dublin, Ireland, representing one of the world\u2019s most extensive conferences on research and technology of spoken language understanding and processing. Experts in speech-related research fields gather to take part in oral presentations and poster sessions and to build collaborations across the globe.\n<\/p>\n

<\/p>\n

\nWe are excited to be a Platinum Sponsor<\/a> of INTERSPEECH 2023, where we will be showcasing more than 20 research publications and supporting a number of workshops and special sessions. We welcome in-person attendees to drop by the Google Research booth to meet our researchers and participate in Q&As and demonstrations of some of our latest speech technologies, which help to improve accessibility and provide convenience in communication for billions of users. In addition, online attendees are encouraged to visit our virtual booth in Topia<\/a> where you can get up-to-date information on research and opportunities at Google. Visit the @GoogleAI<\/a> Twitter account to find out about Google booth activities (e.g., demos and Q&A sessions). You can also learn more about the Google research being presented at INTERSPEECH 2023 below (Google affiliations in bold<\/strong>).\n<\/p>\n

<\/p>\n

Board and Organizing Committee<\/h2>\n
\n

\nISCA Board, Technical Committee Chair: Bhuvana Ramabhadran<\/em><\/strong>\n<\/p>\n

\nArea Chairs include:
\u00a0\u00a0\u00a0\u00a0Analysis of Speech and Audio Signals: Richard Rose<\/em><\/strong>
\u00a0\u00a0\u00a0\u00a0Speech Synthesis and Spoken Language Generation: Rob Clark<\/em><\/strong>
\u00a0\u00a0\u00a0\u00a0Special Areas: Tara Sainath<\/em><\/strong>\n<\/p>\n<\/div>\n

<\/p>\n

Satellite events<\/h2>\n

<\/p>\n

Keynote talk \u2013 ISCA Medalist<\/h2>\n

<\/p>\n

Survey Talk<\/h2>\n
\n

\nSpeech Compression in the AI Era
Speaker: Jan Skoglund<\/em>
<\/strong>\n<\/p>\n<\/div>\n

<\/p>\n

Special session papers<\/h2>\n
\n

\nCascaded Encoders for Fine-Tuning ASR Models on Overlapped Speech<\/a>
Richard Rose<\/b>, Oscar Chang<\/b>, Olivier Siohan<\/b><\/em>\n<\/p>\n

\nTokenSplit: Using Discrete Speech Representations for Direct, Refined, and Transcript-Conditioned Speech Separation and Recognition<\/a>
Hakan Erdogan<\/b>, Scott Wisdom<\/b>, Xuankai Chang*, Zal\u00e1n Borsos<\/b>, Marco Tagliasacchi<\/b>, Neil Zeghidour<\/b>, John R. Hershey<\/b><\/em><\/p>\n<\/div>\n

<\/p>\n

Papers<\/h2>\n
\n

\nDeePMOS: Deep Posterior Mean-Opinion-Score of Speech<\/a>
Xinyu Liang, Fredrik Cumlin, Christian Sch\u00fcldt<\/strong>, Saikat Chatterjee
<\/strong><\/em>\n<\/p>\n

\nO-1: Self-Training with Oracle and 1-Best Hypothesis<\/a>
Murali Karthick Baskar<\/b>, Andrew Rosenberg<\/b>, Bhuvana Ramabhadran<\/b>, Kartik Audhkhasi<\/b><\/em><\/p>\n

\nRe-investigating the Efficient Transfer Learning of Speech Foundation Model Using Feature Fusion Methods<\/a>
Zhouyuan Huo<\/b>, Khe Chai Sim<\/b>, Dongseong Hwang<\/b>, Tsendsuren Munkhdalai<\/b>, Tara N. Sainath<\/b>, Pedro Moreno<\/b><\/em><\/p>\n

\nMOS vs. AB: Evaluating Text-to-Speech Systems Reliably Using Clustered Standard Errors<\/a>
Joshua Camp<\/b>, Tom Kenter<\/b>, Lev Finkelstein<\/b>, Rob Clark<\/b><\/em><\/p>\n

\nLanSER: Language-Model Supported Speech Emotion Recognition<\/a>
Taesik Gong, Josh Belanich<\/strong>, Krishna Somandepalli<\/strong>, Arsha Nagrani<\/strong>, Brian Eoff<\/strong>, Brendan Jou
<\/strong><\/em>\n<\/p>\n

\nModular Domain Adaptation for Conformer-Based Streaming ASR<\/a>
Qiujia Li<\/b>, Bo Li<\/b>, Dongseong Hwang<\/b>, Tara N. Sainath<\/b>, Pedro M. Mengibar<\/b><\/em>\n<\/p>\n

\nOn Training a Neural Residual Acoustic Echo Suppressor for Improved ASR<\/a>
Sankaran Panchapagesan<\/b>, Turaj Zakizadeh Shabestary<\/b>, Arun Narayanan<\/b>
<\/em>\n<\/p>\n

\nMD3: The Multi-dialect Dataset of Dialogues<\/a>
Jacob Eisenstein<\/b>, Vinodkumar Prabhakaran<\/b>, Clara Rivera<\/b>, Dorottya Demszky, Devyani Sharma
<\/em>\n<\/p>\n

\nDual-Mode NAM: Effective Top-K Context Injection for End-to-End ASR<\/a>
Zelin Wu<\/b>, Tsendsuren Munkhdalai<\/b>, Pat Rondon<\/b>, Golan Pundak<\/b>, Khe Chai Sim<\/b>, Christopher Li<\/b>
<\/em>\n<\/p>\n

\nUsing Text Injection to Improve Recognition of Personal Identifiers in Speech<\/a>
Yochai Blau<\/b>, Rohan Agrawal<\/b>, Lior Madmony<\/b>, Gary Wang<\/b>, Andrew Rosenberg<\/b>, Zhehuai Chen<\/b>, Zorik Gekhman<\/b>, Genady Beryozkin<\/b>, Parisa Haghani<\/b>, Bhuvana Ramabhadran<\/b>
<\/em>\n<\/p>\n

\nHow to Estimate Model Transferability of Pre-trained Speech Models?<\/a>
Zih-Ching Chen, Chao-Han Huck Yang*, Bo Li<\/strong>, Yu Zhang<\/strong>, Nanxin Chen<\/strong>, Shuo-yiin Chang<\/strong>, Rohit Prabhavalkar, <\/strong>Hung-yi Lee, Tara N. Sainath
<\/strong><\/em>\n<\/p>\n

\nImproving Joint Speech-Text Representations Without Alignment<\/a>
Cal Peyser<\/b>, Zhong Meng<\/b>, Ke Hu<\/b>, Rohit Prabhavalkar<\/b>, Andrew Rosenberg<\/b>, Tara N. Sainath<\/b>, Michael Picheny, Kyunghyun Cho
<\/em>\n<\/p>\n

\nText Injection for Capitalization and Turn-Taking Prediction in Speech Models<\/a>
Shaan Bijwadia<\/b>, Shuo-yiin Chang<\/b>, Weiran Wang<\/b>, Zhong Meng<\/b>, Hao Zhang<\/b>, Tara N. Sainath<\/b>
<\/em>\n<\/p>\n

\nStreaming Parrotron for On-Device Speech-to-Speech Conversion<\/a>
Oleg Rybakov<\/b>, Fadi Biadsy<\/b>, Xia Zhang<\/b>, Liyang Jiang<\/b>, Phoenix Meadowlark<\/b>, Shivani Agrawal<\/b>
<\/em>\n<\/p>\n

\nSemantic Segmentation with Bidirectional Language Models Improves Long-Form ASR<\/a>
W. Ronny Huang<\/strong>, Hao Zhang<\/strong>, Shankar Kumar<\/strong>, Shuo-yiin Chang<\/strong>, Tara N. Sainath
<\/strong>\n<\/p>\n

\nUniversal Automatic Phonetic Transcription into the International Phonetic Alphabet<\/a>
Chihiro Taguchi, Yusuke Sakai, Parisa Haghani<\/strong>, David Chiang
<\/em>\n<\/p>\n

\nMixture-of-Expert Conformer for Streaming Multilingual ASR<\/a>
Ke Hu<\/b>, Bo Li<\/b>, Tara N. Sainath<\/b>, Yu Zhang<\/b>, Francoise Beaufays<\/b>
<\/em>\n<\/p>\n

\nReal Time Spectrogram Inversion on Mobile Phone<\/a>
Oleg Rybakov<\/b>, Marco Tagliasacchi<\/b>, Yunpeng Li<\/b>, Liyang Jiang<\/b>, Xia Zhang<\/b>, Fadi Biadsy<\/b>
<\/em>\n<\/p>\n

\n2-Bit Conformer Quantization for Automatic Speech Recognition<\/a>
Oleg Rybakov<\/b>, Phoenix Meadowlark<\/b>, Shaojin Ding<\/b>, David Qiu<\/b>, Jian Li<\/b>, David Rim<\/b>, Yanzhang He<\/b>
<\/em>\n<\/p>\n

\nLibriTTS-R: A Restored Multi-speaker Text-to-Speech Corpus<\/a>
Yuma Koizumi<\/b>, Heiga Zen<\/b>, Shigeki Karita<\/b>, Yifan Ding<\/b>, Kohei Yatabe, Nobuyuki Morioka<\/b>, Michiel Bacchiani<\/b>, Yu Zhang<\/b>, Wei Han<\/b>, Ankur Bapna<\/b>
<\/em>\n<\/p>\n

\nPronScribe: Highly Accurate Multimodal Phonemic Transcription from Speech and Text<\/a>
Yang Yu<\/b>, Matthew Perez*, Ankur Bapna<\/b>, Fadi Haik<\/b>, Siamak Tazari<\/b>, Yu Zhang<\/b>
<\/em>\n<\/p>\n

\nLabel Aware Speech Representation Learning for Language Identification<\/a>
Shikhar Vashishth<\/b>, Shikhar Bharadwaj<\/b>, Sriram Ganapathy<\/b>, Ankur Bapna<\/b>, Min Ma<\/b>, Wei Han<\/b>, Vera Axelrod<\/b>, Partha Talukdar<\/b>
<\/em>\n<\/p>\n<\/div>\n

<\/p>\n

\n

\n *<\/b>\u00a0Work done while at Google<\/span><\/p>\n<\/div>\n