Deepgram Aura-2

Overview

Best for: Enterprise voice agents already on Deepgram STT; high concurrency on pay-as-you-go; self-hosting needs.

At a glance

Sub-200 ms vendor claim. $30/1M PAYG, $27/1M Growth; Aura-1 is half price. Concurrency 45 on PAYG (REST and WSS combined). Socket max 60 minutes, 2,400 chars per minute per socket, 20 flushes per minute. HIPAA and SOC 2 are Deepgram vendor claims, not re-verified.

Features

• input streaming (Speak + Flush)
• Clear message for barge-in
• self-hosted deployment option
• telephony encodings

Pricing

Free tier: $200 credit for new accounts

Source: deepgram.com

Setup

1. Create an API key in the Deepgram console.
2. Connect to wss://api.deepgram.com/v1/speak with model, encoding and sample_rate query params.
3. Send {type:'Speak', text} messages as tokens arrive, then {type:'Flush'} at end of turn.
4. Write binary frames as audio; JSON frames are control/metadata (Flushed, Warning).

Endpoint

wss://api.deepgram.com/v1/speak

Authentication

Authorization: Token <API_KEY> header

// npm i ws
import WebSocket from "ws";
import fs from "fs";

const url = "wss://api.deepgram.com/v1/speak?model=aura-2-thalia-en&encoding=linear16&sample_rate=24000";
const ws = new WebSocket(url, { headers: { Authorization: `Token ${process.env.DEEPGRAM_API_KEY}` } });
const out = fs.createWriteStream("out_24k_s16le.pcm");

ws.on("open", () => {
  for (const t of ["Hello there. ", "This text is streamed ", "sentence by sentence."]) {
    ws.send(JSON.stringify({ type: "Speak", text: t }));
  }
  ws.send(JSON.stringify({ type: "Flush" }));
});
ws.on("message", (data, isBinary) => {
  if (isBinary) return out.write(data);
  const msg = JSON.parse(data.toString());
  if (msg.type === "Flushed") ws.send(JSON.stringify({ type: "Close" }));
});
ws.on("close", () => out.end());

Written from the current docs. Check the vendor's SDK version before you ship.

Warnings

Plus 12 warnings that apply to all text-to-speech, streaming APIs. See category warnings.

Limits

• 2,000 characters per Speak payload (413 above that)
• 2,400 characters per minute throughput per socket
• 20 Flush messages per 60 s
• 60-minute max socket lifetime
• Voice and output settings fixed per connection
• TTS concurrency: PAYG 45, Growth 60 (REST + WSS combined)

Models and products

Docs and sources

Docs

• Streaming TTS
• Media output settings
• TTS changelog

Sources used

• deepgram.com/pricing
• developers.deepgram.com/docs/streaming-text-to-speech
• developers.deepgram.com/docs/tts-media-output-settings
• deepgram.com/learn/aura-2-now-speaks-dutch-french-german-italian-japanese