FR3.I: Large Vocabulary Continuous Speech Recognition and Search |
| Session Type: Poster |
| Time: Friday, 8 May, 15:15 - 17:15 |
| Location: On-Demand |
| Session Chairs: Gakuto Kurata, IBM Research - Tokyo and Steve Renals, University of Edinburgh
|
| |
| FR3.I.1: IMPROVING PROPER NOUN RECOGNITION IN END-TO-END ASR BY CUSTOMIZATION OF THE MWER LOSS CRITERION |
| Cal Peyser; Google, Inc. |
| Tara Sainath; Google, Inc. |
| Golan Pundak; Google, Inc. |
| |
| FR3.I.2: NEURAL LATTICE SEARCH FOR SPEECH RECOGNITION |
| Rao Ma; Shanghai Jiao Tong University |
| Hao Li; Shanghai Jiao Tong University |
| Qi Liu; Shanghai Jiao Tong University |
| Lu Chen; Shanghai Jiao Tong University |
| Kai Yu; Shanghai Jiao Tong University |
| |
| FR3.I.3: DELIBERATION MODEL BASED TWO-PASS END-TO-END SPEECH RECOGNITION |
| Ke Hu; Google, Inc. |
| Tara Sainath; Google, Inc. |
| Ruoming Pang; Google, Inc. |
| Rohit Prabhavalkar; Google, Inc. |
| |
| FR3.I.4: ALIGNMENT-LENGTH SYNCHRONOUS DECODING FOR RNN TRANSDUCER |
| George Saon; IBM Research AI |
| Zoltan Tuske; IBM Research AI |
| Kartik Audhkhasi; IBM Research AI |
| |
| FR3.I.5: INCORPORATING WRITTEN DOMAIN NUMERIC GRAMMARS INTO END-TO-END CONTEXTUAL SPEECH RECOGNITION SYSTEMS FOR IMPROVED RECOGNITION OF NUMERIC SEQUENCES |
| Ben Haynor; Google, Inc. |
| Petar Aleksic; Google, Inc. |
| |
| FR3.I.6: LSTM-BASED ONE-PASS DECODER FOR LOW-LATENCY STREAMING |
| Javier Jorge; Universitat Politècnica de València |
| Adrià Giménez; Universitat Politècnica de València |
| Javier Iranzo-Sánchez; Universitat Politècnica de València |
| Joan Albert Silvestre-Cerdà; Universitat Politècnica de València |
| Jorge Civera; Universitat Politècnica de València |
| Albert Sanchis; Universitat Politècnica de València |
| Alfons Juan; Universitat Politècnica de València |
| |
| FR3.I.7: MULTISTATE ENCODING WITH END-TO-END SPEECH RNN TRANSDUCER NETWORK |
| Zelin Wu; Google LLC |
| Bo Li; Google LLC |
| Yu Zhang; Google LLC |
| Petar Aleksic; Google LLC |
| Tara Sainath; Google LLC |
| |
| FR3.I.8: NEURAL ORACLE SEARCH ON N-BEST HYPOTHESES |
| Ehsan Variani; Google |
| Tongzhou Chen; Google |
| James Apfel; Google |
| Bhuvana Ramabhadran; Google |
| Seungji Lee; Google |
| Pedro Moreno; Google |
| |
| FR3.I.10: TRANSFORMER TRANSDUCER: A STREAMABLE SPEECH RECOGNITION MODEL WITH TRANSFORMER ENCODERS AND RNN-T LOSS |
| Qian Zhang; Google |
| Han Lu; Google |
| Hasim Sak; Google |
| Anshuman Tripathi; Google |
| Erik McDermott; Google |
| Stephen Koo; Google |
| Shankar Kumar; Google |
| |
| FR3.I.11: FULL-SUM DECODING FOR HYBRID HMM BASED SPEECH RECOGNITION USING LSTM LANGUAGE MODEL |
| Wei Zhou; RWTH Aachen University |
| Ralf Schlüter; RWTH Aachen University |
| Hermann Ney; RWTH Aachen University |
| |
| FR3.I.12: THE RWTH ASR SYSTEM FOR TED-LIUM RELEASE 2: IMPROVING HYBRID HMM WITH SPECAUGMENT |
| Wei Zhou; RWTH Aachen University |
| Wilfried Michel; RWTH Aachen University |
| Kazuki Irie; RWTH Aachen University |
| Markus Kitza; RWTH Aachen University |
| Ralf Schlüter; RWTH Aachen University |
| Hermann Ney; RWTH Aachen University |
| |