{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PMTOIQTMBV2RYMJ2DOTG3KUN6H","short_pith_number":"pith:PMTOIQTM","schema_version":"1.0","canonical_sha256":"7b26e4426c0d751c313a1ba66daa8df1f20b486fb1dc2e5ccf2a94ddb9d3423a","source":{"kind":"arxiv","id":"2308.16692","version":2},"attestation_state":"computed","paper":{"title":"SpeechTokenizer: Unified Speech Tokenizer for Speech Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Dong Zhang, Shimin Li, Xin Zhang, Xipeng Qiu, Yaqian Zhou","submitted_at":"2023-08-31T12:53:09Z","abstract_excerpt":"Current speech large language models build upon discrete speech representations, which can be categorized into semantic tokens and acoustic tokens. However, existing speech tokens are not specifically designed for speech language modeling. To assess the suitability of speech tokens for building speech language models, we established the first benchmark, SLMTokBench. Our results indicate that neither semantic nor acoustic tokens are ideal for this purpose. Therefore, we propose SpeechTokenizer, a unified speech tokenizer for speech large language models. SpeechTokenizer adopts the Encoder-Decod"},"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":"2308.16692","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-08-31T12:53:09Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"6a1988a33ed5f266f2073d03467b12e5b85c67f43359356cdf61361242fa147e","abstract_canon_sha256":"733a50eff75d9e9afb1ba189fe85812b6e88e963096efe7dc794bff993d99f8b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:36:23.499665Z","signature_b64":"+cluaCY7K+p010S7+yzWqDvuofH7mPUM9mRZqqRva/Z9M3ewrCXZ5UaPX3yT0VPeL95js+Y13FRZn2K3go/wDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b26e4426c0d751c313a1ba66daa8df1f20b486fb1dc2e5ccf2a94ddb9d3423a","last_reissued_at":"2026-07-05T07:36:23.499170Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:36:23.499170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SpeechTokenizer: Unified Speech Tokenizer for Speech Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Dong Zhang, Shimin Li, Xin Zhang, Xipeng Qiu, Yaqian Zhou","submitted_at":"2023-08-31T12:53:09Z","abstract_excerpt":"Current speech large language models build upon discrete speech representations, which can be categorized into semantic tokens and acoustic tokens. However, existing speech tokens are not specifically designed for speech language modeling. To assess the suitability of speech tokens for building speech language models, we established the first benchmark, SLMTokBench. Our results indicate that neither semantic nor acoustic tokens are ideal for this purpose. Therefore, we propose SpeechTokenizer, a unified speech tokenizer for speech large language models. SpeechTokenizer adopts the Encoder-Decod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.16692","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/2308.16692/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":"2308.16692","created_at":"2026-07-05T07:36:23.499231+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.16692v2","created_at":"2026-07-05T07:36:23.499231+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.16692","created_at":"2026-07-05T07:36:23.499231+00:00"},{"alias_kind":"pith_short_12","alias_value":"PMTOIQTMBV2R","created_at":"2026-07-05T07:36:23.499231+00:00"},{"alias_kind":"pith_short_16","alias_value":"PMTOIQTMBV2RYMJ2","created_at":"2026-07-05T07:36:23.499231+00:00"},{"alias_kind":"pith_short_8","alias_value":"PMTOIQTM","created_at":"2026-07-05T07:36:23.499231+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26824","citing_title":"wav2tok 2.0: Scalable Audio Tokenization Maintaining Explicit Pairwise Token Alignment for Efficient Audio Retrieval","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20993","citing_title":"Phonemes to the Rescue: Multilingual Tokenization Based on International Phonetic Alphabet","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12940","citing_title":"Self-Guidance: Enhancing Neural Codecs via Decoder Manifold Alignment","ref_index":86,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12199","citing_title":"Which Speech Representation Better Matches Text-Native Reasoning? A Study of Speech-Text Alignment on Frame Rate and Representation","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11611","citing_title":"SARA: A Dual-Stream VAE for High-Fidelity Speech Generation via Integrating Semantic and Acoustic Representations","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01802","citing_title":"MOSS-Audio Technical Report","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31247","citing_title":"FlexiSLM: A Dynamic and Controllable Frame Rate Spoken Language Model","ref_index":193,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06582","citing_title":"PairAlign: A Framework for Sequence Tokenization via Self-Alignment with Applications to Audio Tokenization","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20519","citing_title":"Codec-Robust Attacks on Audio LLMs","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2512.01537","citing_title":"Two-Dimensional Quantization for Geometry-Aware Audio Coding","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20519","citing_title":"Codec-Robust Attacks on Audio LLMs","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2509.22220","citing_title":"StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs","ref_index":91,"is_internal_anchor":false},{"citing_arxiv_id":"2510.06201","citing_title":"TokenChain: A Discrete Speech Chain via Semantic Token Modeling","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2410.06885","citing_title":"F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching","ref_index":154,"is_internal_anchor":false},{"citing_arxiv_id":"2507.16632","citing_title":"Step-Audio 2 Technical Report","ref_index":84,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06582","citing_title":"PairAlign: A Framework for Sequence Tokenization via Self-Alignment with Applications to Audio Tokenization","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08125","citing_title":"PolySLGen: Online Multimodal Speaking-Listening Reaction Generation in Polyadic Interaction","ref_index":93,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H","json":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H.json","graph_json":"https://pith.science/api/pith-number/PMTOIQTMBV2RYMJ2DOTG3KUN6H/graph.json","events_json":"https://pith.science/api/pith-number/PMTOIQTMBV2RYMJ2DOTG3KUN6H/events.json","paper":"https://pith.science/paper/PMTOIQTM"},"agent_actions":{"view_html":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H","download_json":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H.json","view_paper":"https://pith.science/paper/PMTOIQTM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.16692&json=true","fetch_graph":"https://pith.science/api/pith-number/PMTOIQTMBV2RYMJ2DOTG3KUN6H/graph.json","fetch_events":"https://pith.science/api/pith-number/PMTOIQTMBV2RYMJ2DOTG3KUN6H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H/action/storage_attestation","attest_author":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H/action/author_attestation","sign_citation":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H/action/citation_signature","submit_replication":"https://pith.science/pith/PMTOIQTMBV2RYMJ2DOTG3KUN6H/action/replication_record"}},"created_at":"2026-07-05T07:36:23.499231+00:00","updated_at":"2026-07-05T07:36:23.499231+00:00"}