Deepgram

Deepgram is an enterprise voice AI platform that provides speech-to-text, text-to-speech, and voice agent APIs powered by advanced AI models. The platform offers real-time and batch transcription through its Nova model family, natural-sounding speech synthesis through its Aura model family, and an end-to-end Voice Agent API that combines STT, LLM orchestration, and TTS into a single real-time interface.

5 APIs 16 Features

Artificial IntelligenceSpeech-To-TextText-To-SpeechTranscriptionVoice AI

APIs

Deepgram Speech-To-Text API

The Deepgram Speech-to-Text API provides accurate, fast transcription of audio content using advanced AI models. It supports both pre-recorded audio files and real-time streamin...

Deepgram Text-To-Speech API

The Deepgram Text-to-Speech API converts text into natural-sounding speech using the Aura model family. It supports both single text requests and continuous streaming text-to-sp...

Deepgram Voice Agent API

The Deepgram Voice Agent API is an end-to-end solution that combines speech-to-text, LLM orchestration, and text-to-speech into a single real-time API. It simplifies the develop...

Deepgram Audio Intelligence API

The Deepgram Audio Intelligence API provides advanced analysis capabilities for audio and text content. It offers features including sentiment analysis, summarization, topic det...

Deepgram Management API

The Deepgram Management API allows developers to programmatically manage their Deepgram account resources. It provides endpoints for creating and managing API keys, configuring ...

Collections

Deepgram Management API

POSTMAN

Deepgram Speech-to-Text API

POSTMAN

Deepgram Text-to-Speech API

POSTMAN

Deepgram Management API

OPEN

Deepgram Speech-to-Text API

OPEN

Deepgram Text-to-Speech API

OPEN

Arazzo Workflows

Deepgram Audit Project Usage and Billing

Pull a project's details, usage summary, request log, and billing balances into a single audit snapshot.

ARAZZO

Deepgram Balance-Gated Transcription

Check a project's billing balance and only transcribe audio when sufficient credit remains.

ARAZZO

Deepgram Diarized Transcript Intelligence

Transcribe a multi-speaker recording with diarization, then run targeted text intelligence over the transcript.

ARAZZO

Deepgram Invite and Confirm a Project Member

Send a project invitation, confirm it appears in the pending invitation list, and list current members.

ARAZZO

Deepgram Provision a Project API Key

Create a new project, mint a scoped API key for it, and verify the key appears in the project key list.

ARAZZO

Deepgram Rotate a Project API Key

Mint a replacement API key in a project, verify it, then delete the old key to complete a rotation.

ARAZZO

Deepgram Select a Model and Transcribe

Browse available models, read the metadata for a chosen model, then transcribe audio with that model.

ARAZZO

Deepgram Transcribe, Analyze, and Synthesize

Transcribe audio to text, run text intelligence on the transcript, then synthesize a spoken response.

ARAZZO

Deepgram Transcribe Audio and Track Usage

Transcribe a pre-recorded audio URL and then reconcile the request against project usage and request logs.

ARAZZO

Deepgram Update a Member's Scopes

Locate a project member, read their current scopes, update them, and confirm the new scopes took effect.

ARAZZO

GraphQL

Deepgram GraphQL Schema

This document describes the conceptual GraphQL schema for the Deepgram AI voice platform, covering speech-to-text transcription, text-to-speech synthesis, voice agent capabiliti...

GRAPHQL

Pricing Plans

Deepgram Plans Pricing

6 plans

PLANS

Rate Limits

Deepgram Rate Limits

4 limits

RATE LIMITS

FinOps

Deepgram Finops

FINOPS

Features

Nova-3 STT: $0.0048/min mono, $0.0058/min multilingual

Flux STT: $0.0065/min English, $0.0078/min multilingual

Aura-1 TTS at $0.015/1k characters

Aura-2 TTS at $0.030/1k characters with studio quality

Streaming and pre-recorded transcription

Speaker diarization, smart formatting

Default 50 streaming concurrent (PAYG), 100 pre-recorded

Voice cloning on Aura models

Voice agents combining STT + LLM + TTS

30+ language support on multilingual models

WebSocket streaming for real-time STT

REST API for pre-recorded files

Audio Intelligence: summarization, topics, sentiment, entities

Custom model training (Enterprise)

Self-hosted on-prem option (Enterprise)

OAuth 2.0 + API keys

Sources

opencollection: 1.0.0
info:
  name: Deepgram Text-to-Speech API
  version: '1.0'
request:
  auth:
    type: bearer
    token: '{{bearerToken}}'
items:
- info:
    name: Text-To-Speech
    type: folder
  items:
  - info:
      name: Deepgram Convert text to speech
      type: http
    http:
      method: POST
      url: https://api.deepgram.com/v1/speak
      params:
      - name: model
        value: ''
        type: query
        description: Voice model to use for speech synthesis. Aura model family voices include aura-asteria-en, aura-luna-en,
          aura-stella-en, aura-athena-en, aura-hera-en, aura-orion-en, aura-arcas-en, aura-perseus-en, aura-angus-en, aura-orpheus-en,
          aura-helios-en, and aura-zeus-en.
      - name: encoding
        value: ''
        type: query
        description: Audio encoding format for the output. Options include linear16, mulaw, alaw, mp3, opus, flac, and aac.
      - name: container
        value: ''
        type: query
        description: File format container for the output audio. The default depends on the encoding selected.
      - name: sample_rate
        value: ''
        type: query
        description: Sample rate in Hertz for the output audio. Default is 24000. Valid values depend on the selected encoding.
      - name: bit_rate
        value: ''
        type: query
        description: Bit rate for compressed audio formats such as MP3 and Opus.
      - name: callback
        value: ''
        type: query
        description: URL to which Deepgram will send the audio when processing is complete.
      - name: callback_method
        value: ''
        type: query
        description: HTTP method for the callback request.
      body:
        type: json
        data: '{}'
    docs: Converts text content into natural-sounding speech audio using Deepgram's Aura model family. Returns audio data
      in the specified encoding format. Supports multiple voices and configurable audio output settings including encoding,
      sample rate, and container format.
bundled: true