> ## Documentation Index
> Fetch the complete documentation index at: https://docs.corti.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Create Transcript
> Create a transcript from an audio file attached, via `/recordings` endpoint, to the interaction.
Each interaction may have more than one audio file and transcript associated with it. While audio files up to 60min in total duration, or 150MB in total size, may be attached to an interaction, synchronous processing is only supported for audio files less than ~2min in duration.
If an audio file takes longer to transcribe than the 25sec synchronous processing timeout, then it will continue to process asynchronously. In this scenario, an incomplete or empty transcript with `status=processing` will be returned with a location header that can be used to retrieve the final transcript.
The client can poll the Get Transcript endpoint (`GET /interactions/{id}/transcripts/{transcriptId}/status`) for transcript status changes:
- `200 OK` with status `processing`, `completed`, or `failed`
- `404 Not Found` if the `interactionId` or `transcriptId` are invalid
The completed transcript can be retrieved via the Get Transcript endpoint (`GET /interactions/{id}/transcripts/{transcriptId}/`).
## OpenAPI
````yaml /api-reference/auto-generated-openapi.yml post /interactions/{id}/transcripts/
openapi: 3.0.0
info:
title: Corti API
version: 2.0.0
servers:
- url: https://api.{environment}.corti.app/v2/
variables:
environment:
default: eu
enum:
- us
- eu
security:
- AuthorizationHeader:
- bearer
tags:
- name: Interactions
- name: Facts
- name: Documents
- name: Codes
- name: Recordings
- name: Transcripts
paths:
/interactions/{id}/transcripts/:
post:
tags:
- Transcripts
summary: Create Transcript
description: >-
Create a transcript from an audio file attached, via `/recordings`
endpoint, to the interaction.
Each interaction may have more
than one audio file and transcript associated with it. While audio files
up to 60min in total duration, or 150MB in total size, may be attached
to an interaction, synchronous processing is only supported for audio
files less than ~2min in duration.
If an audio file takes
longer to transcribe than the 25sec synchronous processing timeout, then
it will continue to process asynchronously. In this scenario, an
incomplete or empty transcript with `status=processing` will be returned
with a location header that can be used to retrieve the final
transcript.
The client can poll the Get Transcript endpoint
(`GET /interactions/{id}/transcripts/{transcriptId}/status`) for
transcript status changes:
- `200 OK` with status `processing`,
`completed`, or `failed`
- `404 Not Found` if the `interactionId` or
`transcriptId` are invalid
The completed transcript can be
retrieved via the Get Transcript endpoint (`GET
/interactions/{id}/transcripts/{transcriptId}/`).
operationId: transcripts_create
parameters:
- $ref: '#/components/parameters/Tenant-Name'
- $ref: '#/components/parameters/CommonInteractionId'
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/TranscriptsCreateRequest'
required: true
responses:
'201':
description: >-
Returns the generated transcript, including participant roles and
timestamps for each utterance.
content:
application/json:
schema:
$ref: '#/components/schemas/TranscriptsResponse'
'400':
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: RFC9457
'401':
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: RFC9457
'403':
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: RFC9457
'500':
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: RFC9457
'504':
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
description: RFC9457
x-codeSamples:
- lang: csharp
label: C# .NET SDK
source: |
using Corti;
var client = new CortiClient(
"TENANT_NAME",
CortiClientEnvironment.Eu,
new CortiClientAuth.ClientCredentials("client_id", "client_secret")
);
await client.Transcripts.CreateAsync(
"f47ac10b-58cc-4372-a567-0e02b2c3d479",
new TranscriptsCreateRequest
{
RecordingId = "f47ac10b-58cc-4372-a567-0e02b2c3d479",
PrimaryLanguage = "en",
}
);
- lang: javascript
label: JavaScript SDK
source: >
import { CortiClient, CortiEnvironment } from "@corti/sdk";
const client = new CortiClient({
environment: CortiEnvironment.Eu,
auth: {
clientId: "YOUR_CLIENT_ID",
clientSecret: "YOUR_CLIENT_SECRET"
},
tenantName: "YOUR_TENANT_NAME"
});
await
client.transcripts.create("f47ac10b-58cc-4372-a567-0e02b2c3d479", {
recordingId: "f47ac10b-58cc-4372-a567-0e02b2c3d479",
primaryLanguage: "en"
});
components:
parameters:
Tenant-Name:
name: Tenant-Name
in: header
description: >-
Identifies a distinct entity within Corti's multi-tenant system. Ensures
correct routing and authentication of the request.
required: true
example: base
schema:
type: string
description: >-
Identifies a distinct entity within Corti's multi-tenant system.
Ensures correct routing and authentication of the request.
example: base
CommonInteractionId:
name: id
in: path
description: The unique identifier of the interaction. Must be a valid UUID.
required: true
schema:
$ref: '#/components/schemas/UUID'
schemas:
TranscriptsCreateRequest:
type: object
required:
- recordingId
- primaryLanguage
properties:
recordingId:
$ref: '#/components/schemas/UUID'
type: string
format: uuid
description: The unique identifier for the recording.
items: {}
primaryLanguage:
type: string
example: en
description: >-
The primary spoken language of the recording. Check
https://docs.corti.ai/about/languages for more.
isDictation:
type: boolean
description: >-
Indicates whether spoken dictation commands should be converted to
punctuation (e.g., 'comma' → ',').
isMultichannel:
type: boolean
description: If true, each audio channel is transcribed separately.
diarize:
type: boolean
description: >-
If true, separates speakers within an audio channel returning
incrementing ids for transcript segments.
participants:
type: array
description: >-
An array of participants, each specifying a role and an assigned
audio channel in the recording. Leave empty when shouldDiarize: true
items:
$ref: '#/components/schemas/TranscriptsParticipant'
TranscriptsResponse:
type: object
required:
- id
- metadata
- transcripts
- usageInfo
- recordingId
- status
properties:
id:
$ref: '#/components/schemas/UUID'
type: string
format: uuid
description: The unique identifier of the transcript.
items: {}
metadata:
$ref: '#/components/schemas/TranscriptsMetadata'
type: object
description: >-
Additional information about the participants involved in the
transcript.
transcripts:
type: array
description: An array of transcripts.
nullable: true
items:
$ref: '#/components/schemas/CommonTranscriptResponse'
usageInfo:
$ref: '#/components/schemas/CommonUsageInfo'
type: object
recordingId:
$ref: '#/components/schemas/UUID'
type: string
format: uuid
description: The unique identifier for the associated recording.
items: {}
status:
$ref: '#/components/schemas/TranscriptsStatusEnum'
description: The current status of the transcript processing.
ErrorResponse:
type: object
required:
- requestid
- status
- type
- detail
properties:
requestid:
type: string
status:
type: integer
type:
type: string
detail:
type: string
validationErrors:
type: array
items:
type: object
additionalProperties:
type: string
UUID:
type: string
items: {}
format: uuid
example: f47ac10b-58cc-4372-a567-0e02b2c3d479
TranscriptsParticipant:
type: object
required:
- channel
- role
properties:
channel:
type: integer
description: The audio channel to associate with a participant role.
role:
$ref: '#/components/schemas/TranscriptsParticipantRoleEnum'
description: >-
The role of the participant (e.g., 'doctor', 'patient', use
'multiple' for single channel).
TranscriptsMetadata:
type: object
properties:
participantsRoles:
type: array
items:
$ref: '#/components/schemas/TranscriptsParticipant'
nullable: true
CommonTranscriptResponse:
type: object
required:
- channel
- participant
- speakerId
- text
- start
- end
properties:
channel:
type: integer
description: The channel associated with this phrase/utterance.
participant:
type: integer
description: The identifier of the participant.
speakerId:
type: integer
description: Id to tag an identified speaker. Auto-increments.
text:
type: string
description: The spoken phrase or utterance extracted from the audio.
start:
type: integer
description: Start time in milliseconds for phrase/utterance.
end:
type: integer
description: End time in milliseconds for phrase/utterance.
CommonUsageInfo:
type: object
description: Credits consumed for this request.
required:
- creditsConsumed
properties:
creditsConsumed:
type: number
TranscriptsStatusEnum:
type: string
description: Possible values for transcript processing status.
enum:
- completed
- processing
- failed
TranscriptsParticipantRoleEnum:
type: string
enum:
- doctor
- patient
- multiple
securitySchemes:
AuthorizationHeader:
type: http
description: Input your token
scheme: bearer
````