deepgram
Package
Section titled “Package”@ai-sdk/deepgram
Factories
Section titled “Factories”AIProviders.deepgram
Capabilities
Section titled “Capabilities”- Transcription
- Speech
Deepgram transcription and speech mirror upstream 3.1.7: DEEPGRAM_API_KEY, Token auth, custom headers, versioned user-agent suffix, transcription/transcriptionModel and speech/speechModel aliases, /v1/listen raw-audio transcription with explicit-only diarization and providerOptions.deepgram query mapping, /v1/speak JSON TTS with output format to encoding/container/sample-rate mapping, Aura family plus voice/language model-ID composition, speech speed, incompatible audio cleanup and unsupported-instruction warnings, response-header usage/provider metadata, structured err_code/err_msg errors, and unsupported language/embedding/image families. ProviderV4, ESM-only packaging, Node 22 engines, experimental streaming transcription types, optional JS headers config, and workflow serialization helpers are JS-only upstream concerns.
Factory argument requirements are defined by the public Swift factory signatures. Use Public symbols when you need the exact initializer or factory declaration.
Return to the provider matrix.