> ## Documentation Index > Fetch the complete documentation index at: https://docs.slng.ai/llms.txt > Use this file to discover all available pages before exploring further. # Nova 3 > Transcribe audio using Deepgram Nova 3 with VAD and speaker diarization. **Hosting: routed through SLNG (Deepgram).** This model runs on Deepgram's servers; SLNG proxies each request through the nearest regional gateway. Available in Netherlands, Germany, United States (East, West), Australia, Brazil, United Kingdom, Indonesia, Israel, India, Japan, Singapore, South Africa. Send requests to your region's gateway base URL: `https://eu-north.api.slng.ai`, `https://eu-west.api.slng.ai`, `https://us-east.api.slng.ai`, `https://us-west.api.slng.ai`, `https://au.api.slng.ai`, `https://br.api.slng.ai`, `https://gb.api.slng.ai`, `https://id.api.slng.ai`, `https://il.api.slng.ai`, `https://in.api.slng.ai`, `https://jp.api.slng.ai`, `https://sg.api.slng.ai`, `https://za.api.slng.ai`. ## OpenAPI ````yaml api-reference/openapi/stt-deepgram.oas.yaml POST /v1/stt/deepgram/nova:3 openapi: 3.0.3 info: title: SLNG Gateway API - Deepgram STT version: 0.1.0 description: Unified API for speech-to-text and text-to-speech services. contact: name: SLNG Support email: support@slng.ai servers: - url: https://eu-north.api.slng.ai description: 'Regional gateway: European Union (North) (eu-north)' - url: https://eu-west.api.slng.ai description: 'Regional gateway: European Union (West) (eu-west)' - url: https://us-east.api.slng.ai description: 'Regional gateway: United States (East) (us-east)' - url: https://us-west.api.slng.ai description: 'Regional gateway: United States (West) (us-west)' - url: https://au.api.slng.ai description: 'Regional gateway: Australia (au)' - url: https://br.api.slng.ai description: 'Regional gateway: Brazil (br)' - url: https://gb.api.slng.ai description: 'Regional gateway: United Kingdom (gb)' - url: https://id.api.slng.ai description: 'Regional gateway: Indonesia (id)' - url: https://il.api.slng.ai description: 'Regional gateway: Israel (il)' - url: https://in.api.slng.ai description: 'Regional gateway: India (in)' - url: https://jp.api.slng.ai description: 'Regional gateway: Japan (jp)' - url: https://sg.api.slng.ai description: 'Regional gateway: Singapore (sg)' - url: https://za.api.slng.ai description: 'Regional gateway: South Africa (za)' security: - BearerAuth: [] tags: - name: Deepgram Nova 3 description: Deepgram Nova 3 with VAD and speaker diarization. - name: Deepgram Nova 3 Medical description: Deepgram Nova 3 Medical with specialized vocabulary. paths: /v1/stt/deepgram/nova:3: post: tags: - Deepgram Nova 3 summary: Nova 3 description: Transcribe audio using Deepgram Nova 3 with VAD and speaker diarization. operationId: deepgram/nova:3 requestBody: required: true content: multipart/form-data: schema: $ref: '#/components/schemas/DeepgramNova3Request' examples: binary: summary: Binary audio upload value: audio: language: en application/json: schema: $ref: '#/components/schemas/DeepgramNova3Request' examples: url: summary: URL-based transcription value: url: https://docs.slng.ai/audio/hello.wav language: en responses: '200': $ref: '#/components/responses/DeepgramSttSuccess' '400': $ref: '#/components/responses/ProviderBadRequest' '401': $ref: '#/components/responses/ProviderUnauthorized' '413': $ref: '#/components/responses/ProviderPayloadTooLarge' '500': $ref: '#/components/responses/ProviderInternalServerError' '503': $ref: '#/components/responses/ProviderServiceUnavailable' components: schemas: DeepgramNova3Request: allOf: - $ref: '#/components/schemas/SttAudioRequestBase' - $ref: '#/components/schemas/DeepgramSttOptionsBase' - type: object properties: language: $ref: '#/components/schemas/DeepgramNova3Language' SttAudioRequestBase: type: object properties: audio: type: string format: binary description: Audio file (multipart) or base64-encoded audio (JSON). url: type: string description: Publicly accessible audio URL. example: https://docs.slng.ai/audio/hello.wav DeepgramSttOptionsBase: type: object description: Deepgram STT processing options. properties: model: type: string description: Model name override. default: nova-3 enable_partials: type: boolean description: Enable partial/interim transcription results for streaming. default: true diarize: type: boolean description: >- Enable speaker diarization. Words assigned speaker numbers starting at 0. default: false punctuate: type: boolean description: Add punctuation and capitalization. default: false smart_format: type: boolean description: >- Apply formatting to improve readability (dates, times, numbers, etc.). default: false utterances: type: boolean description: Segment transcript into utterances. default: false paragraphs: type: boolean description: Add paragraph formatting. default: false numerals: type: boolean description: Convert spoken numbers to digits. default: false profanity_filter: type: boolean description: Filter profanity from transcript. default: false redact: type: array description: Redact sensitive information (pci, ssn, numbers). items: type: string enum: - pci - ssn - numbers detect_language: type: boolean description: Auto-detect spoken language. default: false filler_words: type: boolean description: Include filler words (uh, um) in transcript. default: false multichannel: type: boolean description: Enable multi-channel audio processing. default: false keywords: type: array description: Keywords to boost recognition accuracy. items: type: string encoding: type: string description: Input audio encoding. default: linear16 enum: - linear16 - flac - mulaw - amr-nb - amr-wb - opus - speex - mp3 - mp4 - webm - aac - ogg sample_rate: type: integer description: Input audio sample rate in Hz. default: 16000 channels: type: integer description: Number of audio channels. minimum: 1 endpointing: oneOf: - type: integer minimum: 0 - type: boolean default: 10 description: >- Time in milliseconds of silence before finalizing speech. Enabled by default at 10ms on Deepgram's side. Set to false to disable endpointing entirely. utterance_end_ms: type: integer minimum: 200 maximum: 5000 default: 1000 description: >- Time in milliseconds of silence between words before sending an UtteranceEnd message. Default is 1000ms, minimum 200ms, maximum 5000ms. Requires interim_results (enable_partials) to be true. tag: type: string description: Tag for request tracking. DeepgramNova3Language: type: string description: Supported language code. example: multi enum: - multi - es - ru - en - ja - fr - en-in - en-gb - en-nz - nl - en-ca - it - en-us - en-au - de - pt - hi DeepgramSttResponse: type: object description: 3rd Party Deepgram API response format. required: - results properties: results: type: object properties: channels: type: array items: type: object properties: alternatives: type: array items: type: object properties: transcript: type: string confidence: type: number detected_language: type: string metadata: type: object properties: request_id: type: string model: type: string ProviderErrorResponse: type: object description: > Provider error response. Contains error information from the upstream provider. Common errors include invalid speakers, unsupported languages, or malformed requests. properties: error: type: string description: >- Error type or category (e.g., "TTS service error", "Validation error"). example: TTS service error status: type: number description: >- Original HTTP status code from the provider (may differ from gateway response code). example: 400 details: type: string description: > Detailed error message from the provider. For TTS errors, this typically includes the specific validation failure (e.g., invalid speaker ID for the selected model/language). example: >- Rime TTS API error: 400 Bad Request - Invalid argument: Invalid speaker: aurelie required: - error responses: DeepgramSttSuccess: description: Transcription successful. content: application/json: schema: $ref: '#/components/schemas/DeepgramSttResponse' examples: basic: summary: Basic transcription value: results: channels: - alternatives: - transcript: hello from sunny barcelona confidence: 0.98 metadata: request_id: req-123 model: nova-3 ProviderBadRequest: description: Bad request - invalid parameters provided. content: application/json: schema: $ref: '#/components/schemas/ProviderErrorResponse' examples: validation-error: summary: Validation error value: error: Validation error details: 'Missing required field: text' ProviderUnauthorized: description: Invalid or missing API key. content: application/json: schema: $ref: '#/components/schemas/ProviderErrorResponse' ProviderPayloadTooLarge: description: Audio file exceeds size limit. content: application/json: schema: $ref: '#/components/schemas/ProviderErrorResponse' ProviderInternalServerError: description: Internal server error from the provider. content: application/json: schema: $ref: '#/components/schemas/ProviderErrorResponse' examples: processing-error: summary: Processing error value: error: Service error details: Internal processing error ProviderServiceUnavailable: description: Service unavailable. content: application/json: schema: $ref: '#/components/schemas/ProviderErrorResponse' securitySchemes: BearerAuth: type: http scheme: bearer description: | API key issued by SLNG. Pass as `Authorization: Bearer `. ````