{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4MIIFTEIOKWABOMHS5IIOLFBP7","short_pith_number":"pith:4MIIFTEI","schema_version":"1.0","canonical_sha256":"e31082cc8872ac00b9879750872ca17fc90e056f74a348bfcb408c713f3d7179","source":{"kind":"arxiv","id":"2107.03312","version":1},"attestation_state":"computed","paper":{"title":"SoundStream: An End-to-End Neural Audio Codec","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Ahmed Omran, Alejandro Luebs, Jan Skoglund, Marco Tagliasacchi, Neil Zeghidour","submitted_at":"2021-07-07T15:45:42Z","abstract_excerpt":"We present SoundStream, a novel neural audio codec that can efficiently compress speech, music and general audio at bitrates normally targeted by speech-tailored codecs. SoundStream relies on a model architecture composed by a fully convolutional encoder/decoder network and a residual vector quantizer, which are trained jointly end-to-end. Training leverages recent advances in text-to-speech and speech enhancement, which combine adversarial and reconstruction losses to allow the generation of high-quality audio content from quantized embeddings. By training with structured dropout applied to q"},"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":"2107.03312","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2021-07-07T15:45:42Z","cross_cats_sorted":["cs.LG","eess.AS"],"title_canon_sha256":"cca3a7af2254b19f1b8a5a13fe6e6e34943412ee98358360b0045fab7cc891d8","abstract_canon_sha256":"21341a3e244d0f26c1163a5b7f550b519045f212342ff82acfb57a8e71d6aeea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:56:02.657063Z","signature_b64":"X6TZmq9vpRWzOmuLGthhkPw6dZtllT1Mlb2+iXo72FJM0jsjm4FEBTR6CJ4IbtEtzIY/SdVg7ofi2xeivtE+Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e31082cc8872ac00b9879750872ca17fc90e056f74a348bfcb408c713f3d7179","last_reissued_at":"2026-07-05T02:56:02.656632Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:56:02.656632Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SoundStream: An End-to-End Neural Audio Codec","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Ahmed Omran, Alejandro Luebs, Jan Skoglund, Marco Tagliasacchi, Neil Zeghidour","submitted_at":"2021-07-07T15:45:42Z","abstract_excerpt":"We present SoundStream, a novel neural audio codec that can efficiently compress speech, music and general audio at bitrates normally targeted by speech-tailored codecs. SoundStream relies on a model architecture composed by a fully convolutional encoder/decoder network and a residual vector quantizer, which are trained jointly end-to-end. Training leverages recent advances in text-to-speech and speech enhancement, which combine adversarial and reconstruction losses to allow the generation of high-quality audio content from quantized embeddings. By training with structured dropout applied to q"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.03312","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/2107.03312/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":"2107.03312","created_at":"2026-07-05T02:56:02.656698+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.03312v1","created_at":"2026-07-05T02:56:02.656698+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.03312","created_at":"2026-07-05T02:56:02.656698+00:00"},{"alias_kind":"pith_short_12","alias_value":"4MIIFTEIOKWA","created_at":"2026-07-05T02:56:02.656698+00:00"},{"alias_kind":"pith_short_16","alias_value":"4MIIFTEIOKWABOMH","created_at":"2026-07-05T02:56:02.656698+00:00"},{"alias_kind":"pith_short_8","alias_value":"4MIIFTEI","created_at":"2026-07-05T02:56:02.656698+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21893","citing_title":"AugCodec: A Low-Bitrate Disentangled Neural Speech Codec via Data Augmentation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02266","citing_title":"HERMES: A Multi-Granularity Labeling Substrate for Pre-training Data Mixtures","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06357","citing_title":"F3-Tokenizer: Taming Audio Autoencoder Latents for Understanding and Generation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02739","citing_title":"EntangleCodec: A Unified Discrete Audio Tokenizer via Semantic-Acoustic Entanglement","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02631","citing_title":"Wavelet as Tokenizer: Preliminary Results on a Shared Wavelet Token Schema for Natural Signals","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28779","citing_title":"Telephony Voice Agent for Banking Services","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20519","citing_title":"Codec-Robust Attacks on Audio LLMs","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20519","citing_title":"Codec-Robust Attacks on Audio LLMs","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20649","citing_title":"AMAR: Lightweight Attention-Based Multi-User Activity Recognition from Wi-Fi CSI","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21081","citing_title":"Musical Attention Transformer: Music Generation Using a Music-Specific Attention Model","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13789","citing_title":"ENSEMBITS: an alphabet of protein conformational ensembles","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13789","citing_title":"ENSEMBITS: an alphabet of protein conformational ensembles","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2604.01929","citing_title":"Woosh: A Sound Effects Foundation Model","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12965","citing_title":"Efficient Retrieval Scaling with Hierarchical Indexing for Large Scale Recommendation","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2501.09747","citing_title":"FAST: Efficient Action Tokenization for Vision-Language-Action Models","ref_index":69,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7","json":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7.json","graph_json":"https://pith.science/api/pith-number/4MIIFTEIOKWABOMHS5IIOLFBP7/graph.json","events_json":"https://pith.science/api/pith-number/4MIIFTEIOKWABOMHS5IIOLFBP7/events.json","paper":"https://pith.science/paper/4MIIFTEI"},"agent_actions":{"view_html":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7","download_json":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7.json","view_paper":"https://pith.science/paper/4MIIFTEI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.03312&json=true","fetch_graph":"https://pith.science/api/pith-number/4MIIFTEIOKWABOMHS5IIOLFBP7/graph.json","fetch_events":"https://pith.science/api/pith-number/4MIIFTEIOKWABOMHS5IIOLFBP7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7/action/storage_attestation","attest_author":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7/action/author_attestation","sign_citation":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7/action/citation_signature","submit_replication":"https://pith.science/pith/4MIIFTEIOKWABOMHS5IIOLFBP7/action/replication_record"}},"created_at":"2026-07-05T02:56:02.656698+00:00","updated_at":"2026-07-05T02:56:02.656698+00:00"}