{ "vercy": "1.0-draft", "publication": { "status": "research-draft", "adjudicationStatus": "unreviewed", "publishableCanonical": false, "generatedAt": "2026-09-12T09:12:59Z", "providers": [ "Codex" ], "breadth": "recalled by Codex without web access - no source was read", "missingProviders": [], "pass": 2, "cost": { "grok": { "seconds": 23.9, "error": "Reading additional input from stdin...\nOpenAI Codex v0.153.4\n--------\nworkdir: R:\\02_PROJECTS\\02_Meta_Models_Platforms\\Ver.cy\\current\\thing-registry-backlog\nmodel: gpt-6-astra\nprovider: openai\napproval: never\nsandbox: read-only\nreasoning effort: none\nreasoning summaries: none\nsession id: 01a094e3-0993-7d72-a1b6-572036fc93f0\n--------\nuser\nDescribe what is already known about one registered thing. Answer as JSON only, no prose around it.\n\nThing: diphone\nSense to describe: (none recorded)\nDomain code: ACT.PRC\nAlso known as: (none)\n\n\nContext for this batch of 1001 things:\n# Batch 043: 1001 registe", "usd": 0.0, "recall": true }, "codex": { "seconds": 77.7, "error": "Reading additional input from stdin...\nOpenAI Codex v0.153.4\n--------\nworkdir: R:\\02_PROJECTS\\02_Meta_Models_Platforms\\Ver.cy\\current\\thing-registry-backlog\nmodel: gpt-6-astra\nprovider: openai\napproval: never\nsandbox: read-only\nreasoning effort: none\nreasoning summaries: none\nsession id: 01a094e3-0984-7a32-aa75-10cc69d447f7\n--------\nuser\nYou are drafting a Vercy meta-model for one registered thing. Answer as JSON only, no prose around it.\n\nThing: diphone\nRegistry id: vr.tr.diphone\nPlane / domain: ACT / ACT.PRC\nRegistry definition: (none recorded)\nNames folded into this entry: (none)\n\n\nContext " } }, "enrichment": { "pass": 3, "engine": "claude", "producedAt": "2026-10-06T12:00:00Z", "ingestedAt": "2026-10-06T08:16:20+00:00", "fields": [ "agent_conduct", "ethics", "failure_modes_and_hazards", "owners" ] }, "gate": { "passed": true, "version": "1.0", "at": "2026-10-07T10:10:25+00:00", "evidence": [ "completeness 1.00", "sense: Wikidata description", "2 live sources", "review by codex: pass" ] } }, "metaModel": { "id": "THING-Q1227145", "registryId": "vr.tr.diphone", "name": "diphone", "version": "0.1.0-research.1", "entryKind": "thing", "family": "Thing Registry", "domain": [ "ACT.PRC" ], "status": "research-draft" }, "canonicalUrl": "https://ver.cy/models/thing/q1227145/", "model": { "registry_id": "vr.tr.diphone", "name": "diphone", "purpose": "Enable an agent to identify a diphone, assess whether a recorded realization captures the intended phone transition, and judge its suitability for speech synthesis.", "definition": "A diphone is a speech unit spanning approximately the middle of one phone to the middle of the next, capturing their transition and commonly serving as a recorded building block in concatenative speech synthesis.", "scope_statement": "This model owns the ordered two-phone transition represented by a diphone, its operational segmentation, recorded realizations, and conditions for selection and concatenation; it distinguishes the abstract transition type from individual audio tokens.", "in_scope": [ "Ordered phone-pair identity within a declared language variety and phone inventory", "Operational definition of the segment spanning the transition between adjacent phones", "Recorded tokens, segmentation landmarks, and extraction provenance", "Contextual variants and coverage of required transitions", "Compatibility, transformation limits, and quality of diphone joins" ], "out_of_scope": [ "Complete phoneme inventories and phonological theories", "Diphthongs as vowel categories and their internal trajectories", "Whole-word pronunciation and text-to-phone conversion", "Complete speech corpora and recording-session management", "Synthesizer architecture and utterance-level prosody generation" ], "distinguishing_features": [ "A synthesis diphone conventionally extends from an interior point of one phone to an interior point of the next, preserving their intervening transition; it is not simply two complete phone recordings. See [Festival waveform synthesis notes](https://www.cs.cmu.edu/~awb/11752/notes/festtut_7.html).", "Its identity is ordered: a transition from phone A to phone B does not imply equivalence to B followed by A.", "It requires two adjacent phone positions under a declared segmentation; the two positions may carry the same phone label.", "It differs from a diphthong: diphthong status concerns vowel organization, whereas diphone status concerns a transition between two phone positions.", "It differs from a triphone representation, which identifies a central phone together with left and right context." ], "characteristics": [ { "name": "Ordered phone identity", "kind": "relation", "unit_or_values": "First phone and second phone linked to a named, versioned inventory", "why_it_matters": "Makes transition identity interpretable and prevents accidental reversal or mixing of transcription conventions." }, { "name": "Representational level", "kind": "category", "unit_or_values": "Transition type; recorded token; transformed token", "why_it_matters": "Separates the linguistic target from a particular recording or processed realization." }, { "name": "Segmentation convention", "kind": "category", "unit_or_values": "Temporal midpoint; selected stable region; implementation-specific convention", "why_it_matters": "Determines which acoustic interval counts as the diphone." }, { "name": "Temporal landmarks", "kind": "measurement", "unit_or_values": "Start, internal phone boundary, and end in milliseconds or sample indices", "why_it_matters": "Supports reproducible extraction and separate inspection of the two phone portions." }, { "name": "Phonetic context", "kind": "relation", "unit_or_values": "Language variety, allophones, stress, syllable position, adjacent context, and speaking style", "why_it_matters": "Explains why tokens with the same basic pair labels may differ in suitability." }, { "name": "Acoustic realization", "kind": "measurement", "unit_or_values": "Duration in milliseconds; voiced-region F0 in hertz; spectral representation with declared method", "why_it_matters": "Supports comparison with target timing, pitch, and neighboring units without assigning pitch to unvoiced regions." }, { "name": "Target-relative usability", "kind": "state", "unit_or_values": "Unassessed; suitable; suitable with constraints; unsuitable, with evidence and target conditions", "why_it_matters": "Avoids treating a diphone as universally usable after one successful synthesis." } ], "affordances": [ "Resolve a requested phone transition to matching diphone types and candidate tokens.", "Extract or revise a token using documented phone boundaries and join landmarks.", "Identify missing required transitions within a specified pronunciation workload.", "Compare contextual variants and select a token for a synthesis target.", "Assess concatenation and proposed pitch or duration changes under declared quality criteria.", "Reject, relabel, or request rerecording of a token when transition identity or acoustic quality is inadequate." ], "agent_conduct": { "may": [ "Build and use diphone inventories from recordings made with the speaker's consent.", "Explain diphone synthesis and its limits from technical sources." ], "must_not": [ "Record or reuse a person's voice to build a synthetic voice without their consent.", "Use synthetic speech to impersonate a real person.", "Present synthesized speech as a real recording.", "Use a voice beyond the scope its donor agreed to." ], "requires_human": [ "Releasing a synthetic voice built from an identifiable person." ] }, "ethics": { "considerations": [ "A voice is part of a person's identity; copying it without consent is a wrong.", "Synthetic voices give speech to people who have lost their own, and should serve them well." ], "affected_parties": [ "Voice donors", "People who rely on synthetic speech", "Listeners who can be deceived" ] }, "owners": { "steward": "The person whose voice was recorded, with the developer who holds the inventory under their agreement.", "master_systems": [] }, "failure_modes_and_hazards": [ "Synthetic voices can be misused for fraud and impersonation.", "Poor joins between units make speech hard to understand for users who depend on it." ], "what_it_is_for": [ "Diphones are used in speech technology to represent the transition between adjacent phones.", "In diphone-based concatenative synthesis, recorded units are selected and joined to produce spoken utterances.", "They are also examined when assessing pronunciation coverage and the quality of transitions in recorded speech inventories." ], "facets_not_applicable": { "interfaces": "No generally applicable standard or official classification specifically governing diphones is identified.", "recognition": "A diphone is a speech unit identified through phonetic and acoustic analysis rather than visual appearance." } }, "sources": [ { "id": "wikidata", "title": "Wikidata item Q1227145: diphone", "url": "https://www.wikidata.org/wiki/Q1227145", "url_status": "live", "what_it_supports": "identity and sense of the item", "checked_by": "wikidata-api" }, { "id": "wikipedia-en", "title": "Wikipedia: Diphone", "url": "https://en.wikipedia.org/wiki/Diphone", "url_status": "live", "what_it_supports": "general description of the item", "checked_by": "wikidata-api sitelink" } ], "structure": { "bundles": [ { "id": "transition-identity", "name": "Transition identity", "description": "Establish what the diphone denotes before interpreting its labels or audio.", "rationale": "Pair labels alone do not establish phone meaning, adjacency, or the distinction between a type and a token.", "layers": [ { "id": "definition-and-instance", "name": "Definition and instance", "description": "Make the adopted sense and instance criteria explicit.", "findings": [ { "id": "adopted-diphone-sense", "name": "Adopted diphone sense", "description": "Record whether the application uses diphone for an adjacent-phone transition type or for a bounded acoustic realization, and identify the authority for that usage.", "questions": [ { "text": "What definition does the relevant linguistic reference or synthesis system adopt, and what qualifies as an instance?", "kind": "definition", "id": "adopted-diphone-sense-q01" }, { "text": "Does this record describe a transition type, an original audio token, or a transformed token?", "kind": "boundary", "id": "adopted-diphone-sense-q02" } ] } ] }, { "id": "ordered-phone-reference", "name": "Ordered phone reference", "description": "Bind both phone positions to an interpretable inventory.", "findings": [ { "id": "phone-label-resolution", "name": "Phone label resolution", "description": "Record ordered labels, inventory version, language variety, and any mapping between phonemic input and realized phones.", "questions": [ { "text": "Which inventory and language variety define the first and second phone labels?", "kind": "provenance", "id": "phone-label-resolution-q01" }, { "text": "Do the labels denote realized phones, phonemes, or synthesis-specific variants, and how are these mapped?", "kind": "definition", "id": "phone-label-resolution-q02" } ] } ] } ] }, { "id": "acoustic-segmentation", "name": "Acoustic segmentation", "description": "Locate the transition and the two outer cuts in a recording.", "rationale": "The internal phone boundary and the outer concatenation boundaries serve different purposes and must not be confused.", "layers": [ { "id": "outer-cut-selection", "name": "Outer cut selection", "description": "Document how interior points of the two phones become diphone endpoints.", "findings": [ { "id": "endpoint-convention", "name": "Endpoint convention", "description": "Distinguish automatic temporal midpoints from deliberately selected cut points. FestVox documents midpoint defaults with explicit boundary overrides: [diphone database construction](https://www.cs.cmu.edu/afs/cs/user/lenzo/html/areas/papers/festvox/festvox_5.html).", "questions": [ { "text": "Were the start and end selected by temporal midpoint, acoustic stability, or another documented rule?", "kind": "measurement", "id": "endpoint-convention-q01" }, { "text": "How should endpoints be selected when a short or rapidly changing phone has no clear stable region?", "kind": "action", "id": "endpoint-convention-q02" } ] } ] }, { "id": "internal-transition-location", "name": "Internal transition location", "description": "Locate the phone boundary without treating coarticulation as an instantaneous event.", "findings": [ { "id": "transition-landmarks", "name": "Transition landmarks", "description": "Record the internal boundary, its uncertainty, and acoustic evidence that the extracted interval retains the intended transition.", "questions": [ { "text": "Where are the start, internal phone boundary, and end on the source recording's time base?", "kind": "measurement", "id": "transition-landmarks-q01" }, { "text": "What acoustic evidence supports the phone boundary, and how much annotation uncertainty remains?", "kind": "measurement", "id": "transition-landmarks-q02" } ] } ] } ] }, { "id": "context-and-coverage", "name": "Context and coverage", "description": "Determine which distinctions and transitions the intended application requires.", "rationale": "An inventory of nominal phone pairs does not establish coverage of the realizations needed for a particular voice or workload.", "layers": [ { "id": "contextual-variants", "name": "Contextual variants", "description": "Represent meaningful differences among tokens sharing basic pair labels.", "findings": [ { "id": "variant-selection-conditions", "name": "Variant selection conditions", "description": "Record when allophonic or contextual variants need separate selection. Festival documents augmented diphone identities, including distinctions between light and dark lateral realizations: [diphone selection](https://www.cstr.ed.ac.uk/projects/festival/manual/festival_21.html).", "questions": [ { "text": "Which stress, allophone, syllable-position, or cluster distinctions change the required realization of this pair?", "kind": "boundary", "id": "variant-selection-conditions-q01" }, { "text": "What evidence permits one token to substitute for another contextual variant?", "kind": "action", "id": "variant-selection-conditions-q02" } ] } ] }, { "id": "required-transition-coverage", "name": "Required transition coverage", "description": "Assess availability against an explicit set of expected phone sequences.", "findings": [ { "id": "coverage-and-exceptions", "name": "Coverage and exceptions", "description": "Separate required, attested, excluded, missing, and convention-dependent pairs, including repeated labels and silence-related units.", "questions": [ { "text": "Which target pronunciations or phonotactic rules establish that this ordered pair is required?", "kind": "provenance", "id": "coverage-and-exceptions-q01" }, { "text": "Does the application include silence, utterance-edge units, and repeated-phone pairs, and which required cases lack usable tokens?", "kind": "boundary", "id": "coverage-and-exceptions-q02" } ] } ] } ] }, { "id": "token-evidence", "name": "Token evidence", "description": "Connect a diphone realization to its recording and establish whether it is trustworthy.", "rationale": "A correctly named interval may still contain an unintended pronunciation, a segmentation error, or damaged audio.", "layers": [ { "id": "recording-and-extraction-lineage", "name": "Recording and extraction lineage", "description": "Preserve evidence needed to reconstruct and interpret a token.", "findings": [ { "id": "recoverable-source-interval", "name": "Recoverable source interval", "description": "Link the token to the source utterance, speaker or voice, elicitation context, annotations, and processing history.", "questions": [ { "text": "Which recording, speaker, elicitation text, and source interval produced this token?", "kind": "provenance", "id": "recoverable-source-interval-q01" }, { "text": "Which labeling, resampling, trimming, or signal-processing steps changed the token or its timing coordinates?", "kind": "provenance", "id": "recoverable-source-interval-q02" } ] } ] }, { "id": "phonetic-and-signal-review", "name": "Phonetic and signal review", "description": "Assess both intended pronunciation and usable signal quality.", "findings": [ { "id": "token-acceptance-evidence", "name": "Token acceptance evidence", "description": "Record review of phone identity, retained transition, clipping, noise, and relevant voicing landmarks before accepting a token.", "questions": [ { "text": "Does listening and acoustic inspection support the intended phones and transition rather than merely the expected transcript?", "kind": "measurement", "id": "token-acceptance-evidence-q01" }, { "text": "Which observed defects require boundary correction, relabeling, exclusion, or rerecording?", "kind": "action", "id": "token-acceptance-evidence-q02" } ] } ] } ] }, { "id": "synthesis-use", "name": "Synthesis use", "description": "Judge whether a token can meet a target and connect acceptably to neighboring units.", "rationale": "Matching the requested pair is necessary but insufficient for successful concatenation and prosodic modification.", "layers": [ { "id": "neighbor-join-compatibility", "name": "Neighbor join compatibility", "description": "Evaluate joins at the shared phone interior.", "findings": [ { "id": "join-suitability", "name": "Join suitability", "description": "Assess whether adjacent diphones share compatible realizations of their common phone and whether their acoustic endpoints produce an acceptable join.", "questions": [ { "text": "Do the neighboring units agree on the shared phone's inventory identity and required contextual realization?", "kind": "boundary", "id": "join-suitability-q01" }, { "text": "What listening results and declared spectral, amplitude, or voicing measures support accepting the join?", "kind": "measurement", "id": "join-suitability-q02" } ] } ] }, { "id": "target-adaptation-and-fallback", "name": "Target adaptation and fallback", "description": "Bound transformations and define responses when no adequate token is available.", "findings": [ { "id": "permitted-token-adaptation", "name": "Permitted token adaptation", "description": "Record target duration and pitch, synthesis-method requirements, tested modification limits, and explicit fallback conditions.", "questions": [ { "text": "What duration and voiced-region pitch changes have been evaluated for this token with the intended synthesis method?", "kind": "measurement", "id": "permitted-token-adaptation-q01" }, { "text": "When a token is absent or fails quality criteria, should the agent select another variant, invoke a documented fallback, or request new recording?", "kind": "action", "id": "permitted-token-adaptation-q02" } ] } ] } ] }, { "id": "evidence-and-external-alignment", "name": "Evidence and external alignment", "description": "What the world already says about this thing, gathered so the model can be checked against it.", "rationale": "A model that cannot be lined up against existing standards, identifiers and practice cannot be adopted by anyone who already uses them.", "layers": [ { "id": "reported-evidence", "name": "Reported evidence", "description": "Findings from the breadth pass, kept separate from the structural claims.", "findings": [ { "id": "evidence-confidence-notes", "name": "Check these first", "description": "Recalled without web access and unsourced; every item is a lead to verify.", "evidence": [ "The speech-technology sense is assumed because no sense was recorded; the ACT.PRC domain assignment should be checked.", "Midpoint boundaries are an idealisation: practical segmentation commonly targets relatively steady portions of each phone.", "Inventory sizes, naming conventions and treatment of silence are implementation-specific; no universal identifier or mandatory diphone standard is asserted." ], "questions": [ { "id": "evidence-confidence-notes-q01", "text": "Which of these check these first hold for the sense of diphone this model covers, and on what evidence?", "kind": "provenance" } ] }, { "id": "evidence-real-world-use", "name": "Real-world use", "description": "Recalled without web access and unsourced; every item is a lead to verify.", "evidence": [ "Diphone speech synthesis concatenates recorded units to generate utterances from a phonetic transcription.", "Recorded diphones can be modified in duration and fundamental frequency to implement target timing and intonation.", "Diphone inventories support research and teaching about speech synthesis and coarticulation." ], "questions": [ { "id": "evidence-real-world-use-q01", "text": "Which of these real-world use hold for the sense of diphone this model covers, and on what evidence?", "kind": "provenance" } ] }, { "id": "evidence-measurements", "name": "Typical measurements", "description": "Recalled without web access and unsourced; every item is a lead to verify.", "evidence": [ { "quantity": "Unit duration", "typical_range": "Depends on the constituent phones, speaker and speaking rate; no universal range.", "unit": "ms" }, { "quantity": "Inventory coverage", "typical_range": "Measured against the phone transitions required by the target language and transcription system.", "unit": "%" } ], "questions": [ { "id": "evidence-measurements-q01", "text": "Which of these typical measurements hold for the sense of diphone this model covers, and on what evidence?", "kind": "provenance" } ] }, { "id": "evidence-failure-modes-and-hazards", "name": "Failure modes and hazards", "description": "Recalled without web access and unsourced; every item is a lead to verify.", "evidence": [ "Incorrect segmentation places joins in acoustically unstable regions, producing audible discontinuities.", "Missing transitions require substitutions that can reduce naturalness or intelligibility.", "A single recording per diphone may inadequately represent variation caused by stress, prosody and surrounding phones.", "Large pitch or duration modifications can introduce audible signal-processing artifacts.", "Inconsistent recording conditions can create changes in loudness or voice quality between concatenated units." ], "questions": [ { "id": "evidence-failure-modes-and-hazards-q01", "text": "Which of these failure modes and hazards hold for the sense of diphone this model covers, and on what evidence?", "kind": "provenance" } ] }, { "id": "evidence-regional-variation", "name": "Regional variation", "description": "Recalled without web access and unsourced; every item is a lead to verify.", "evidence": [ "Language and dialect affect the phone inventory and which transitions a diphone database must cover.", "Pronunciation differences mean a database recorded for one accent may not adequately represent another." ], "questions": [ { "id": "evidence-regional-variation-q01", "text": "Which of these regional variation hold for the sense of diphone this model covers, and on what evidence?", "kind": "provenance" } ] }, { "id": "evidence-neighbours", "name": "Neighbouring kinds and how to tell them apart", "description": "Recalled without web access and unsourced; every item is a lead to verify.", "evidence": [ { "name": "phone", "difference": "A phone is an individual speech sound; a diphone spans portions of two successive phones and their transition." }, { "name": "phoneme", "difference": "A phoneme is a contrastive category in a language; a diphone is a segment defined across adjacent speech sounds." }, { "name": "diphthong", "difference": "A diphthong is a vowel with changing quality within a syllable; a diphone can span any relevant adjacent phone pair." }, { "name": "triphone", "difference": "In common speech-recognition usage, a triphone represents a central phone conditioned on its left and right neighbours; a diphone captures the transition between two phones." }, { "name": "demisyllable", "difference": "A demisyllable is defined relative to syllable structure rather than the midpoints of successive phones." } ], "questions": [ { "id": "evidence-neighbours-q01", "text": "Which of these neighbouring kinds and how to tell them apart hold for the sense of diphone this model covers, and on what evidence?", "kind": "provenance" } ] } ] } ] } ] }, "openQuestions": [ "Does vr.tr.diphone intend the general adjacent-phone concept, the bounded synthesis unit, or both, and what evidence supports its ACT / ACT.PRC classification?", "Which authoritative references establish the relationship between phone-pair usage and acoustic-segment usage across phonetics and speech synthesis?", "Which endpoint rules are defensible for stops, affricates, very short phones, and transitions lacking stable interior regions?", "Which language-specific contextual distinctions require separate diphone identities, and which can remain token selection attributes?", "What target-specific evidence should determine acceptable join quality, transformation limits, and fallback substitutions?" ], "statistics": { "bundles": 6, "layers": 11, "findings": 16, "questions": 26 } }