OSS-first docs

These docs teach the open system first: contracts, generated surfaces, runtimes, governance, and incremental adoption. Studio shows up as the operating layer on top, not as the source of truth.

ai-voice-stt.mistral

Mistral audio transcription integration for batch and streaming speech-to-text use cases.

  • Type: integration (integration)
  • Version: 1.0.0
  • Owners: platform.ai
  • Tags: voice, stt, transcription, mistral, voxtral
  • field.key.label
    ai-voice-stt.mistral
    field.version.label
    1.0.0
    field.type.label
    integration (integration)
    field.title.label
    ai-voice-stt.mistral
    field.description.label

    Mistral audio transcription integration for batch and streaming speech-to-text use cases.

  • Type: integration (integration)
  • Version: 1.0.0
  • Owners: platform.ai
  • Tags: voice, stt, transcription, mistral, voxtral
  • field.tags.label
    voice,stt,transcription,mistral,voxtral
    field.owners.label
    platform.ai
    field.stability.label

    Mistral audio transcription integration for batch and streaming speech-to-text use cases.

    Source Definition

    export const mistralSttIntegrationSpec = defineIntegration({
    	meta: {
    		key: 'ai-voice-stt.mistral',
    		version: '1.0.0',
    		category: 'ai-voice-stt',
    		title: 'Mistral Speech-to-Text',
    		description:
    			'Mistral audio transcription integration for batch and streaming speech-to-text use cases.',
    		domain: 'ai',
    		owners: ['platform.ai'],
    		tags: ['voice', 'stt', 'transcription', 'mistral', 'voxtral'],
    		stability: StabilityEnum.Experimental,
    	},
    	supportedModes: ['managed', 'byok'],
    	transports: [
    		{ type: 'rest', baseUrl: 'https://api.mistral.ai' },
    	] satisfies IntegrationTransportConfig[],
    	preferredTransport: 'rest',
    	supportedAuthMethods: [
    		{ type: 'api-key', headerName: 'Authorization', prefix: 'Bearer ' },
    	] satisfies IntegrationAuthConfig[],
    	capabilities: {
    		provides: [{ key: 'ai.voice.stt', version: '1.0.0' }],
    	},
    	configSchema: {
    		schema: {
    			type: 'object',
    			properties: {
    				model: {
    					type: 'string',
    					description:
    						'Mistral transcription model identifier (e.g. voxtral-mini-latest).',
    				},
    				language: {
    					type: 'string',
    					description: 'Default language code for transcription (e.g. en).',
    				},
    				diarize: {
    					type: 'boolean',
    					description:
    						'Enable speaker diarization when supported by the model.',
    				},
    				wordTimestamps: {
    					type: 'boolean',
    					description: 'Include word-level timestamp metadata when available.',
    				},
    			},
    		},
    		example: {
    			model: 'voxtral-mini-latest',
    			language: 'en',
    			diarize: true,
    			wordTimestamps: true,
    		},
    	},
    	secretSchema: {
    		schema: {
    			type: 'object',
    			required: ['apiKey'],
    			properties: {
    				apiKey: {
    					type: 'string',
    					description:
    						'Mistral API key with access to audio transcription endpoints.',
    				},
    			},
    		},
    		example: {
    			apiKey: 'mistral-***',
    		},
    	},
    	healthCheck: {
    		method: 'custom',
    		timeoutMs: 5000,
    	},
    	docsUrl: 'https://docs.mistral.ai/api/#tag/audio',
    	byokSetup: {
    		setupInstructions:
    			'Generate a Mistral API key and ensure the project has access to Voxtral or another transcription-capable model.',
    		keyRotationSupported: true,
    		quotaTrackingSupported: true,
    	},
    });