> ## Documentation Index > Fetch the complete documentation index at: https://docs.chariot.in/llms.txt > Use this file to discover all available pages before exploring further. # Create a forced alignment > Align audio of up to 90 seconds with its transcript and get every word's start and end time. ## OpenAPI ````yaml api-reference/openapi.json POST /v1/forced-alignment openapi: 3.1.0 info: title: Chariot API version: '1' description: >- Chariot is a voice AI platform built for Indian and English languages. Convert text into natural-sounding speech over a simple REST call, an HTTP audio stream, or a low-latency WebSocket, and transcribe speech to text over a streaming WebSocket or from a prerecorded file (Beta). This reference documents the public, API-key-authenticated endpoints. contact: name: Chariot Support email: support@chariot.in url: https://chariot.in servers: - url: https://api.chariot.in description: Production security: - APIKeyHeader: [] tags: - name: Text to Speech description: Convert text into speech synchronously or as a stream. - name: Speech to Text description: >- Transcribe prerecorded files (Beta) and review streaming speech-to-text sessions. - name: Forced Alignment description: Time a transcript against its audio, word by word (Beta). - name: Voices description: Browse the voice catalog. - name: Account description: Inspect your credit balance and usage. paths: /v1/forced-alignment: post: tags: - Forced Alignment summary: Create a forced alignment description: >- **Beta.** Align audio of up to 90 seconds with its transcript and receive a start and end time for every word in the same response. Costs 1 credit per second of audio, rounded up. See the [Forced alignment guide](/guides/forced-alignment/overview). operationId: createForcedAlignment requestBody: required: true content: multipart/form-data: schema: $ref: '#/components/schemas/ForcedAlignmentRequest' responses: '200': description: The completed alignment. content: application/json: schema: $ref: '#/components/schemas/ForcedAlignment' example: alignment_id: 6f1c2d9e-3b7a-4f0e-9a51-2c8d4e7b1a90 status: completed language: hi filename: greeting.wav duration_seconds: 1.83 text_characters: 21 word_count: 4 credits_used: 2 loss: 0.06 words: - text: नमस्ते, start: 0.12 end: 0.58 loss: 0.04 - text: आप start: 0.66 end: 0.81 loss: 0.06 - text: कैसे start: 0.81 end: 1.1 loss: 0.05 - text: हैं? start: 1.1 end: 1.42 loss: 0.09 text: नमस्ते, आप कैसे हैं? has_audio: false error: null created_at: '2026-10-29T10:00:00Z' started_at: '2026-10-29T10:00:00Z' completed_at: '2026-10-29T10:00:01Z' content_expires_at: null '401': $ref: '#/components/responses/Unauthorized' '402': description: Not enough credits for this audio. content: application/json: schema: $ref: '#/components/schemas/Error' example: statusCode: 402 error: HTTPException message: >- This audio is 12.0s and needs 12 credits. The workspace does not have enough. '403': description: Forced alignment is not included in the current plan. content: application/json: schema: $ref: '#/components/schemas/Error' '413': description: >- The upload is well over 50 MB, refused while it is still uploading. Nothing is charged. This body has a single `detail` field, not the standard error envelope. A file just over 50 MB returns `422` instead. content: application/json: schema: type: object required: - detail properties: detail: type: string example: detail: 'Request is too large: the file limit is 50 MB' '422': description: >- Invalid input: unreadable audio, too long or too short (under 0.1 s), a file just over 50 MB, an empty transcript, one over 5,000 characters or one with no words that can be aligned, or an unsupported language. Nothing is charged. content: application/json: schema: $ref: '#/components/schemas/Error' '429': description: Too many concurrent requests for this workspace. content: application/json: schema: $ref: '#/components/schemas/Error' example: statusCode: 429 error: HTTPException message: >- Concurrent request limit exceeded (8) for alignment.standard on the current plan '500': description: Something went wrong on Chariot's side. The credits were refunded. content: application/json: schema: $ref: '#/components/schemas/Error' '503': description: The aligner is unavailable. Any charge was refunded; retry later. content: application/json: schema: $ref: '#/components/schemas/Error' '504': description: The alignment timed out. Any charge was refunded; retry later. content: application/json: schema: $ref: '#/components/schemas/Error' security: - APIKeyHeader: [] components: schemas: ForcedAlignmentRequest: type: object required: - file - text properties: file: type: string format: binary description: >- The audio: WAV, MP3, FLAC, OGG, M4A, AAC or WebM. Video containers such as MP4 must contain an audio track. text: type: string description: >- The transcript of the audio as plain text, up to 5,000 characters. Speaker labels and timestamps are not supported. language: type: string default: hi description: >- ISO 639-1 code of the spoken language, for example `hi`, `en`, `ta`. See `GET /v1/forced-alignment/options` for the full list. ForcedAlignment: type: object required: - alignment_id - status - language - filename - duration_seconds - text_characters - credits_used - created_at properties: alignment_id: type: string format: uuid status: type: string enum: - processing - completed - failed language: type: string description: ISO 639-1 code the audio was aligned as. filename: type: - string - 'null' description: >- Name of the uploaded file, for display. `null` after `content_expires_at`. duration_seconds: type: number description: >- Duration of the audio, measured by Chariot. This is what you are billed for. text_characters: type: integer word_count: type: - integer - 'null' credits_used: type: integer description: >- Credits charged: the duration in seconds, rounded up. `0` when a failed alignment was refunded. loss: type: - number - 'null' description: Average alignment loss over the transcript. Lower is better. words: type: - array - 'null' items: $ref: '#/components/schemas/ForcedAlignmentWord' description: >- One entry per whitespace-separated word of your transcript, in order. Present once `status` is `completed`; `null` in lists and before completion. Punctuation-only words have `start` equal to `end`. `null` after `content_expires_at`. text: type: - string - 'null' description: >- The transcript as aligned. Returned when you create the alignment and by its detail; `null` in lists, and once the content has expired. has_audio: type: boolean description: >- Whether the audio you sent is stored and available from `GET /v1/forced-alignment/{alignment_id}/audio`. It is stored just after the alignment is created, so it can read `false` for a moment. `false` once the content has expired. error: oneOf: - $ref: '#/components/schemas/ForcedAlignmentError' - type: 'null' created_at: type: string format: date-time started_at: type: - string - 'null' format: date-time completed_at: type: - string - 'null' format: date-time content_expires_at: type: - string - 'null' format: date-time description: >- When the words, transcript, filename and stored audio expire under your workspace's data retention setting. After it, `words`, `text` and `filename` read as `null` and `has_audio` is `false`; duration, status and credits remain. `null` means kept until deleted. Error: type: object description: Standard error envelope returned by all endpoints. properties: statusCode: type: integer description: HTTP status code, echoed in the body. error: type: string description: Machine-readable error class name. message: type: string description: Human-readable description of what went wrong. required: - statusCode - error - message ForcedAlignmentWord: type: object required: - text - start - end properties: text: type: string description: >- The word, exactly as it appears in your transcript, punctuation included. start: type: number description: Start time in seconds from the beginning of the audio. end: type: number description: End time in seconds from the beginning of the audio. loss: type: - number - 'null' description: >- Alignment loss for this word. Lower means the aligner is more confident. `null` when not available. ForcedAlignmentError: type: object required: - code - message properties: code: type: string enum: - transcript_unalignable - audio_too_short - aligner_unavailable - aligner_timeout - interrupted - internal_error description: >- Why the alignment failed. `transcript_unalignable`: the transcript has no words that can be aligned. `audio_too_short`: the audio is too short to align. `aligner_unavailable`: the aligner could not process the request. `aligner_timeout`: the aligner took too long. `interrupted`: the request was interrupted on Chariot's side. `internal_error`: something went wrong on Chariot's side. message: type: string description: >- What went wrong, in plain words. A generic message once the content has expired. responses: Unauthorized: description: Authentication failed. The API key is missing, invalid, or expired. content: application/json: schema: $ref: '#/components/schemas/Error' examples: missing: summary: No credentials provided value: statusCode: 401 error: UnauthorizedException message: No API key or user credentials provided invalid: summary: Invalid or unknown API key value: statusCode: 401 error: UnauthorizedException message: Invalid API Key or user credentials securitySchemes: APIKeyHeader: type: apiKey in: header name: chariotai-api-key description: >- Your Chariot API key, generated from the dashboard. Send it in the `chariotai-api-key` header on every request. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.