> ## Documentation Index > Fetch the complete documentation index at: https://anam.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # import pronunciation dictionary > Creates a pronunciation dictionary from a Cartesia pronunciation dictionary file: an array of `{ text, pronunciation, case_sensitive? }` rows, with an optional Anam `sounds_like` fallback. A pronunciation containing `<<` or `>>` is imported as an IPA entry; anything else is a sounds-like entry. The file is validated as a whole: it must hold 1 to 500 rows and at most 1 MiB of UTF-8 JSON. When `json` is an array, the limit applies to the array serialized without whitespace. If any row is invalid, nothing is saved and the response lists every error by one-based row and file field name. ## OpenAPI ````yaml https://api.anam.ai/swagger.json post /v1/pronunciation-dictionaries/import openapi: 3.1.0 info: title: Anam AI API version: '1.0' servers: - url: https://api.anam.ai description: Anam API security: - BearerAuth: [] tags: - name: Sessions - name: Avatar Videos - name: Meetings - name: Personas - name: Avatars - name: Voices - name: Pronunciation Dictionaries - name: LLMs - name: Knowledge - name: Tools - name: Share Links - name: Engine paths: /v1/pronunciation-dictionaries/import: post: tags: - Pronunciation Dictionaries summary: import pronunciation dictionary description: > Creates a pronunciation dictionary from a Cartesia pronunciation dictionary file: an array of `{ text, pronunciation, case_sensitive? }` rows, with an optional Anam `sounds_like` fallback. A pronunciation containing `<<` or `>>` is imported as an IPA entry; anything else is a sounds-like entry. The file is validated as a whole: it must hold 1 to 500 rows and at most 1 MiB of UTF-8 JSON. When `json` is an array, the limit applies to the array serialized without whitespace. If any row is invalid, nothing is saved and the response lists every error by one-based row and file field name. operationId: importPronunciationDictionary requestBody: required: true content: application/json: schema: type: object additionalProperties: false required: - name - json properties: name: type: string minLength: 1 maxLength: 200 description: Display name for the new dictionary example: Brand names json: description: >- The file's rows, either as an array or as the file's contents in a string. oneOf: - type: array minItems: 1 maxItems: 500 items: type: object additionalProperties: false required: - text - pronunciation properties: text: type: string maxLength: 100 pronunciation: type: string maxLength: 200 description: >- Respelling, or Cartesia IPA markup between `<<` and `>>` sounds_like: type: string maxLength: 200 description: >- IPA rows only. Respelling spoken by voices whose model can't read IPA. case_sensitive: type: boolean default: false - type: string description: The file's contents examples: default: $ref: '#/components/examples/PronunciationDictionaryImport' responses: '201': description: Successfully imported pronunciation dictionary content: application/json: schema: $ref: '#/components/schemas/PronunciationDictionary' examples: default: $ref: >- #/components/examples/PronunciationDictionaryImportedResponse '400': description: Bad request - The body is not valid JSON content: application/json: schema: $ref: '#/components/schemas/PronunciationDictionaryError' '401': description: Unauthorized - Invalid or missing API key '403': description: Forbidden - API key lacks write access to pronunciation dictionaries '422': description: >- Unprocessable - The name or file is invalid. `field` uses the file's names, such as `case_sensitive`. Nothing was saved. content: application/json: schema: $ref: '#/components/schemas/PronunciationDictionaryValidationError' examples: default: $ref: >- #/components/examples/PronunciationDictionaryImportValidationError '429': description: >- Too many requests - Your organization made more than 60 pronunciation dictionary writes (creates, imports, updates and archives) this minute. Rejected writes count too. headers: Retry-After: schema: type: integer description: Seconds until the limit resets content: application/json: schema: $ref: '#/components/schemas/PronunciationDictionaryError' '500': description: Server error components: examples: PronunciationDictionaryImport: summary: Import Cartesia-format rows value: name: Brand names json: - text: Anam pronunciation: annum - text: SQL pronunciation: sequel case_sensitive: true - text: Zorvex pronunciation: <<ˈ|z|ɔ|ɹ|v|ɛ|k|s>> sounds_like: zor vex PronunciationDictionaryImportedResponse: summary: The imported dictionary value: id: 4f600129-13cf-4bfa-909a-2f4c31467f43 name: Brand names revision: 1 createdAt: '2026-10-11T09:00:00.000Z' updatedAt: '2026-10-11T09:00:00.000Z' entries: - text: Anam kind: sounds_like pronunciation: annum caseSensitive: false - text: SQL kind: sounds_like pronunciation: sequel caseSensitive: true - text: Zorvex kind: ipa pronunciation: <<ˈ|z|ɔ|ɹ|v|ɛ|k|s>> soundsLike: zor vex caseSensitive: false PronunciationDictionaryImportValidationError: summary: Invalid file rows; nothing was saved value: error: Invalid dictionary import details: - row: 2 field: case_sensitive message: Expected boolean, received string schemas: PronunciationDictionary: description: > A pronunciation dictionary and its entries. Saves are live: sessions that start after a save use the saved entries, and running sessions keep the entries they started with. allOf: - $ref: '#/components/schemas/PronunciationDictionarySummary' - type: object properties: entries: type: array items: $ref: '#/components/schemas/PronunciationEntry' PronunciationDictionaryError: type: object properties: error: type: string description: What went wrong code: type: string enum: - REVISION_CONFLICT - DICTIONARY_ATTACHED - RATE_LIMITED description: Why a 409 or 429 happened personaIds: type: array items: type: string format: uuid description: With `DICTIONARY_ATTACHED`, the personas that use the dictionary PronunciationDictionaryValidationError: type: object properties: error: type: string example: Invalid pronunciation dictionary code: type: string enum: - DICTIONARY_OUTDATED description: Set when the saved entries, not the request, fail the current rules details: type: array description: Every validation error items: type: object properties: row: type: - integer - 'null' minimum: 1 description: One-based entry number, or null for an error outside an entry field: type: - string - 'null' description: Field with the error, or null for the whole entry or request message: type: string description: What to fix PronunciationDictionarySummary: type: object properties: id: type: string format: uuid description: >- Pronunciation dictionary ID. Pass it as `voiceGenerationOptions.pronunciationDictionaryId` when creating a session token. example: 4f600129-13cf-4bfa-909a-2f4c31467f43 name: type: string description: Display name example: Brand names revision: type: integer minimum: 1 description: >- Starts at 1 and increases with every save. Send it as `expectedRevision` to reject a write if someone else saved first. example: 1 createdAt: type: string format: date-time description: When the dictionary was created updatedAt: type: string format: date-time description: When the dictionary was last saved PronunciationEntry: type: object additionalProperties: false required: - text - pronunciation description: > One word or phrase and how to say it. The persona speaks the pronunciation in place of the text; captions and transcripts keep the original text. Text and pronunciation must not be blank or contain `[` or `]`. Two entries can't have the same text (ignoring case and spacing) unless both set `caseSensitive` and differ only in case. properties: text: type: string maxLength: 100 description: Word or phrase to replace, at most 100 Unicode characters. example: Anam kind: type: string enum: - sounds_like - ipa default: sounds_like description: > - `sounds_like`: `pronunciation` is a respelling, such as `annum`. Works with every voice. - `ipa`: `pronunciation` is Cartesia IPA markup, such as `<<ˈ|ɑ|n|ə|m>>`. Voices whose model reads IPA (Cartesia Sonic 3, ElevenLabs v3 and v4) speak it; other voices speak `soundsLike`, or leave the text unchanged without it. example: sounds_like pronunciation: type: string maxLength: 200 description: >- How to say the text, at most 200 Unicode characters. A sounds-like respelling can't contain `<<` or `>>`. example: annum soundsLike: type: string maxLength: 200 description: >- IPA entries only. Respelling spoken by voices whose model can't read IPA. example: annum caseSensitive: type: boolean default: false description: Match the text only with the same capitalization. example: false securitySchemes: BearerAuth: type: http scheme: bearer ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.