Aura-2
Deepgram's current TTS model, with 37 English voices plus voices in Spanish, Dutch, German, Italian, French and Japanese - seven languages total.
- Weights
- Not released
- API
- Yes
- Commercial use
- Vendor states commercial use is permitted
It is priced per 1,000 characters and positioned for voice agents and IVR rather than media production. A newer Flux TTS model is also listed on the pricing page.
Standard Deepgram commercial API terms; no voice cloning product documented on the pages fetched, so no clone consent process applies.
Specification
- What it takes in
- text
- What it makes
- audio
- Controls
- streaming-low-latency
- Weights
- Not released
- API
- Yes
- Pricing
- Aura-2: $0.030 per 1,000 characters pay-as-you-go, $0.027 on Growth ($30 and $27 per 1M characters). Aura-1: $0.0150/$0.0135 per 1K. Flux TTS: free until 9/12 then $0.0450/1K (PAYG) or $0.0405/1K (Growth). Nova-3 STT from $0.0048/min streaming monolingual. Voice Agent API $0.056/min through 9/12 then $0.075/min.
- Released or updated
- not established
- Languages
- 7, not individually listed by the vendor
Strengths
- Transparent per-1K-character pricing with a Growth discount
- 80+ voices in total across seven languages, with many English accents
- Same vendor for STT and TTS simplifies agent stacks
- $200 free starting credit reported
Weaknesses
- Only seven languages - weak for localisation work
- No voice cloning
- Latency, sample rate and per-request character limits not documented on the model docs page
- Flux TTS at $0.045/1K will be the most expensive option in the lineup once free access ends
- Aimed at agents, not video VO
Best for
- voice agents and IVR
- English-heavy conversational products
- teams already using Deepgram STT
Where the sources disagree
Recorded rather than resolved. Picking the tidier number would hide the fact that the vendor's own pages do not agree.
- Third-party summaries cite 90ms optimised latency and sub-200ms baseline time-to-first-byte for Aura-2; neither figure appears on the Deepgram model docs page fetched, so latency is recorded as unverified.
- The pricing page lists Flux TTS but the TTS models docs page fetched did not describe it, so Flux specs are unknown.
- '9/12' promotional end dates on the pricing page are given without a year.
What to understand before using it
Explainers naming the mechanisms this model's own documentation mentions.
MCP servers that front Deepgram
Servers in the media MCP directory whose own description or homepage names this vendor. Absence of a server here is not evidence that none exists: most publish no tool list at all.
-
BlitzReels Video Editor
Create, edit, reframe, caption, organize, generate, and export short-form videos with BlitzReels.
See it from the vendor
Shotcraft hosts no sample output. Judging a model means watching its own showreel with your own eyes, and a clip re-hosted here would be a clip we chose.
Nearby in the directory
Azure AI Speech text-to-speech (Neural, Neural HD, MAI voices)
Microsoft's enterprise speech synthesis service, with 200+ prebuilt neural voices across 80+ locales plus tiers for HD voices, multilingual voices, mu
Voice and speech
Sonic 3.5
Cartesia's current TTS model, documented at sub-90ms first-byte latency across 42 languages. It is tuned for conversational delivery and for reading a
Voice and speech
XTTS-v2
An open-weights multilingual TTS and voice cloning model covering 17 languages at 24kHz, still widely used for local voice cloning. Coqui Inc. shut do
Voice and speech