As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/deepgram-flux-tts/

# Deepgram Flux TTS

Deepgram, Flux family. Status: current. Type: speech and audio. Released: 2026-08-12.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Deepgram API (real-time) | `flux-haley-en` | wss://api.deepgram.com/v2/speak?model=flux-haley-en | https://developers.deepgram.com/docs/flux-tts/overview |
| Deepgram API (batch) | `flux-{voice}-en` | https://api.deepgram.com/v2/speak | https://developers.deepgram.com/docs/flux-tts/voices |

## Pricing

- Per 1K characters: $0.045
- Unit as published: per 1K characters pay-as-you-go ($0.0405 Growth); free until 2026-09-12, standard pricing from 2026-09-13
- Source: https://deepgram.com/pricing
- Last checked: 2026-09-29

## Specs

- Input: text
- Output: audio
- Context window: not published
- Max output: not published
- Open weights: no
- Licence: proprietary

## Where it stands

- Newer speech and audio models in the registry since its release: 35. The 5 newest:
  - NVIDIA Nemotron 3.5 Transcribe (30B-A4B) (2026-10-28) https://postcutoff.com/m/nemotron-3-5-transcribe/
  - Hy-SimulMT-30B-A3B (2026-10-22) https://postcutoff.com/m/hy-simulmt-30b-a3b/
  - Eleven Realtime v1 Max / Mini (2026-10-20) https://postcutoff.com/m/eleven-realtime-v1-max/
  - Voxtral Mini Transcribe 3 (+ Mini Transcribe Realtime 3) (2026-10-20) https://postcutoff.com/m/voxtral-mini-transcribe-3/
  - Voxtral Small Transcribe 3 (2026-10-20) https://postcutoff.com/m/voxtral-small-transcribe-3/
- Before it from Deepgram: Deepgram Flux (conversational STT, English + Multilingual) (2025-10-02) https://postcutoff.com/m/deepgram-flux/
- Closed weights; access is hosted only.

## What stands out

- Conversation-native TTS: Keeps context and voice consistency across turns of a conversation instead of treating each sentence in isolation; turn lifecycle events; on Interrupt reports exactly what the user heard (`text_spoken`). Source: https://deepgram.com/learn/text-to-speech-comes-of-age-deepgram-launches-conversation-native-speech
- ~80 ms response, structured-content accuracy: Starts responding in as little as 80 ms under production load; tuned for account numbers, alphanumerics, drug names and money amounts. Source: https://deepgram.com/learn/text-to-speech-comes-of-age-deepgram-launches-conversation-native-speech

## Notes

English only (39 voices; American, British, Irish, Australian, Indian, Singaporean, Filipino accents); use Aura-2 for other languages. Self-hosted GA 2026-08-26; speed 0.5-1.5 and expressivity -2..2 controls. Launched alongside Deepgram passing $100M ARR.

Voice-agent TTS served on /v2/speak (WebSocket for streaming LLM tokens, REST for batch).

Sources: https://developers.deepgram.com/docs/flux-tts/overview , https://developers.deepgram.com/docs/flux-tts/voices , https://deepgram.com/pricing , https://developers.deepgram.com/changelog

Model page: https://postcutoff.com/m/deepgram-flux-tts/