{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:IZH4GIYFIX56665OJ263O42GD7","short_pith_number":"pith:IZH4GIYF","schema_version":"1.0","canonical_sha256":"464fc3230545fbef7bae4ebdb773461fd024e4a605c23efb9406bb667ad7a837","source":{"kind":"arxiv","id":"2006.10296","version":2},"attestation_state":"computed","paper":{"title":"Boosting Objective Scores of a Speech Enhancement Model by MetricGAN Post-processing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Cheng Yu, Chien-Feng Liao, Heng-Cheng Kuo, Kuo-Hsuan Hung, Ryandhimas E. Zezario, Shang-Yi Chuang, Syu-Siang Wang, Szu-Wei Fu, Tsun-An Hsieh, Yen-Ju Lu, You-Jin Li, Yu Tsao","submitted_at":"2020-06-18T06:22:09Z","abstract_excerpt":"The Transformer architecture has demonstrated a superior ability compared to recurrent neural networks in many different natural language processing applications. Therefore, our study applies a modified Transformer in a speech enhancement task. Specifically, positional encoding in the Transformer may not be necessary for speech enhancement, and hence, it is replaced by convolutional layers. To further improve the perceptual evaluation of the speech quality (PESQ) scores of enhanced speech, the L_1 pre-trained Transformer is fine-tuned using a MetricGAN framework. The proposed MetricGAN can be "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2006.10296","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2020-06-18T06:22:09Z","cross_cats_sorted":["cs.LG","cs.SD"],"title_canon_sha256":"c666db915ab10c0b0233b67bc2708b3593300609bf6a7adc5fc1ecdbed713c64","abstract_canon_sha256":"b1727c06be9ebca3a53b23ffe1a64157168e8ffc3b97492f4bc536f272d5d845"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:19:54.778421Z","signature_b64":"eC7DOMHIaWkCeHCw0eukQKYKaPcg1K5lkWaW6w0RJov4Za0GOgQ11Z6feYqpI4WS4IJ7F41/dN2UHMCX66RUAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"464fc3230545fbef7bae4ebdb773461fd024e4a605c23efb9406bb667ad7a837","last_reissued_at":"2026-07-05T02:19:54.777950Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:19:54.777950Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Boosting Objective Scores of a Speech Enhancement Model by MetricGAN Post-processing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Cheng Yu, Chien-Feng Liao, Heng-Cheng Kuo, Kuo-Hsuan Hung, Ryandhimas E. Zezario, Shang-Yi Chuang, Syu-Siang Wang, Szu-Wei Fu, Tsun-An Hsieh, Yen-Ju Lu, You-Jin Li, Yu Tsao","submitted_at":"2020-06-18T06:22:09Z","abstract_excerpt":"The Transformer architecture has demonstrated a superior ability compared to recurrent neural networks in many different natural language processing applications. Therefore, our study applies a modified Transformer in a speech enhancement task. Specifically, positional encoding in the Transformer may not be necessary for speech enhancement, and hence, it is replaced by convolutional layers. To further improve the perceptual evaluation of the speech quality (PESQ) scores of enhanced speech, the L_1 pre-trained Transformer is fine-tuned using a MetricGAN framework. The proposed MetricGAN can be "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.10296","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2006.10296/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2006.10296","created_at":"2026-07-05T02:19:54.778004+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.10296v2","created_at":"2026-07-05T02:19:54.778004+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.10296","created_at":"2026-07-05T02:19:54.778004+00:00"},{"alias_kind":"pith_short_12","alias_value":"IZH4GIYFIX56","created_at":"2026-07-05T02:19:54.778004+00:00"},{"alias_kind":"pith_short_16","alias_value":"IZH4GIYFIX56665O","created_at":"2026-07-05T02:19:54.778004+00:00"},{"alias_kind":"pith_short_8","alias_value":"IZH4GIYF","created_at":"2026-07-05T02:19:54.778004+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7","json":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7.json","graph_json":"https://pith.science/api/pith-number/IZH4GIYFIX56665OJ263O42GD7/graph.json","events_json":"https://pith.science/api/pith-number/IZH4GIYFIX56665OJ263O42GD7/events.json","paper":"https://pith.science/paper/IZH4GIYF"},"agent_actions":{"view_html":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7","download_json":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7.json","view_paper":"https://pith.science/paper/IZH4GIYF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.10296&json=true","fetch_graph":"https://pith.science/api/pith-number/IZH4GIYFIX56665OJ263O42GD7/graph.json","fetch_events":"https://pith.science/api/pith-number/IZH4GIYFIX56665OJ263O42GD7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7/action/storage_attestation","attest_author":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7/action/author_attestation","sign_citation":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7/action/citation_signature","submit_replication":"https://pith.science/pith/IZH4GIYFIX56665OJ263O42GD7/action/replication_record"}},"created_at":"2026-07-05T02:19:54.778004+00:00","updated_at":"2026-07-05T02:19:54.778004+00:00"}