{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:TWASQ7I2VMDIPQYK6UKTFZRTJQ","short_pith_number":"pith:TWASQ7I2","schema_version":"1.0","canonical_sha256":"9d81287d1aab0687c30af51532e6334c1bd5e03c8557ea6c25f3da999b6e7e6f","source":{"kind":"arxiv","id":"2207.07850","version":1},"attestation_state":"computed","paper":{"title":"Reducing Geographic Disparities in Automatic Speech Recognition via Elastic Weight Consolidation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Andreas Stolcke, Brian King, Jasha Droppo, Pegah Ghahremani, Roland Maas, Viet Anh Trinh","submitted_at":"2022-07-16T06:04:52Z","abstract_excerpt":"We present an approach to reduce the performance disparity between geographic regions without degrading performance on the overall user population for ASR. A popular approach is to fine-tune the model with data from regions where the ASR model has a higher word error rate (WER). However, when the ASR model is adapted to get better performance on these high-WER regions, its parameters wander from the previous optimal values, which can lead to worse performance in other regions. In our proposed method, we utilize the elastic weight consolidation (EWC) regularization loss to identify directions i"},"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":"2207.07850","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-07-16T06:04:52Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"5d478a23344cd0950b9db139246cf1816d234a4abd24685bfc008206d57457cb","abstract_canon_sha256":"68853bf15ed82bbd565974cc43636dbd582bded8606a6ba5c53589f975ac9c7d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:42:37.997681Z","signature_b64":"E+Ps00R+/+W5yY9NLkoCLsV6Geby8IYhbN8EOpvlR/aDCE3UYGd+pJOR/MGNa6mO1Ty2oh23xP9Wbv39UqfsBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d81287d1aab0687c30af51532e6334c1bd5e03c8557ea6c25f3da999b6e7e6f","last_reissued_at":"2026-07-05T07:42:37.997259Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:42:37.997259Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reducing Geographic Disparities in Automatic Speech Recognition via Elastic Weight Consolidation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Andreas Stolcke, Brian King, Jasha Droppo, Pegah Ghahremani, Roland Maas, Viet Anh Trinh","submitted_at":"2022-07-16T06:04:52Z","abstract_excerpt":"We present an approach to reduce the performance disparity between geographic regions without degrading performance on the overall user population for ASR. A popular approach is to fine-tune the model with data from regions where the ASR model has a higher word error rate (WER). However, when the ASR model is adapted to get better performance on these high-WER regions, its parameters wander from the previous optimal values, which can lead to worse performance in other regions. In our proposed method, we utilize the elastic weight consolidation (EWC) regularization loss to identify directions i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.07850","kind":"arxiv","version":1},"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/2207.07850/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":"2207.07850","created_at":"2026-07-05T07:42:37.997326+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.07850v1","created_at":"2026-07-05T07:42:37.997326+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.07850","created_at":"2026-07-05T07:42:37.997326+00:00"},{"alias_kind":"pith_short_12","alias_value":"TWASQ7I2VMDI","created_at":"2026-07-05T07:42:37.997326+00:00"},{"alias_kind":"pith_short_16","alias_value":"TWASQ7I2VMDIPQYK","created_at":"2026-07-05T07:42:37.997326+00:00"},{"alias_kind":"pith_short_8","alias_value":"TWASQ7I2","created_at":"2026-07-05T07:42:37.997326+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.10747","citing_title":"FairASR: Fair Audio Contrastive Learning for Automatic Speech Recognition","ref_index":8,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ","json":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ.json","graph_json":"https://pith.science/api/pith-number/TWASQ7I2VMDIPQYK6UKTFZRTJQ/graph.json","events_json":"https://pith.science/api/pith-number/TWASQ7I2VMDIPQYK6UKTFZRTJQ/events.json","paper":"https://pith.science/paper/TWASQ7I2"},"agent_actions":{"view_html":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ","download_json":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ.json","view_paper":"https://pith.science/paper/TWASQ7I2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.07850&json=true","fetch_graph":"https://pith.science/api/pith-number/TWASQ7I2VMDIPQYK6UKTFZRTJQ/graph.json","fetch_events":"https://pith.science/api/pith-number/TWASQ7I2VMDIPQYK6UKTFZRTJQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ/action/storage_attestation","attest_author":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ/action/author_attestation","sign_citation":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ/action/citation_signature","submit_replication":"https://pith.science/pith/TWASQ7I2VMDIPQYK6UKTFZRTJQ/action/replication_record"}},"created_at":"2026-07-05T07:42:37.997326+00:00","updated_at":"2026-07-05T07:42:37.997326+00:00"}