deepgram-speech-core-openapi
Core Deepgram API for transcribing pre-recorded audio, generating speech from text, and listing available models. The user supplies a Deepgram API key at connection time; no credential is included. Official documentation: https://developers.deepgram.com/reference/deepgram-api-overview
Connect this API to an AI agent Browse all specs
What you can do with deepgram-speech-core-openapi via MCP
Every endpoint below becomes a governed tool your AI agent (Claude, ChatGPT, or a customer-facing VX agent) can call — with scoped permissions, sandbox testing, and audit logs.
| Method | Path | Operation | Description |
|---|---|---|---|
| POST | /listen |
transcribeDeepgramAudio | Transcribe pre-recorded audio |
| POST | /speak |
synthesizeDeepgramSpeech | Convert text to speech |
| GET | /models |
listDeepgramModels | List available models |
Authentication
apiKeyAuth
API key — sent as the Authorization header
Provide the Deepgram API key in the Authorization header through the secure connection configuration.
Usage guide
Getting credentials
- Create or sign in to an account at Deepgram.
- Open the Deepgram Console and create an API key.
- When VX Agent asks for authentication, provide the key in the
Authorizationheader using the formatToken YOUR_DEEPGRAM_API_KEY.
Keep the API key private. Do not paste it into the Swagger file, marketplace description, prompts, or chat messages.
How to use this API
Use this API to transcribe audio, analyze speech, generate speech from text, and discover available Deepgram models.
A typical speech-to-text workflow is:
- Use
GET /modelsto review available speech-to-text and text-to-speech models. - Use
POST /listenwith an HTTPS audio or video URL. - Set
model,language, andsmart_formatwhen appropriate. - Enable
diarize,utterances,paragraphs,summarize, orsentimentonly when those results are needed. - Read the returned transcript, words, utterances, speaker information, and analysis results.
A typical text-to-speech workflow is:
- Use
POST /speakwith the text to convert and a supported voice model. - Select an audio encoding such as
mp3,wav, oropus. - Save the generated audio securely and do not expose private content.
Only submit audio that the user owns or is authorized to process. Confirm before processing private meetings, calls, interviews, or a large amount of audio.
Safety and usage notes
Transcripts may contain personal, confidential, or legally sensitive information. Review names, numbers, and important statements before relying on a transcript. Do not expose API keys or private recordings.
Official documentation: Deepgram API Overview.
How to use this spec
- Create a free VX Agents account (sandbox tools included).
- Open the marketplace and connect “deepgram-speech-core-openapi” to an agent.
- Expose it as an MCP server for Claude/ChatGPT, or chat with it directly.