Deepgram Text to Speech
Deepgram Text to Speech reads out the text you give it with a Deepgram Aura-2 voice and returns an audio file: MP3 by default, WAV or OGG if you choose. The voice sets the language: there are Spanish voices and English voices. If the text is long, it splits it at the end of sentences, reads each part and joins them into a single file.
You need a Deepgram API credential with your Deepgram account key. See Credentials. The text is sent to Deepgram to produce the audio.
This is not the voice node for calls: to speak to a caller during a live call, the Voice Agent uses the Deepgram TTS node. This component returns a finished file.
When to use it
- Send your customers a voice note confirming an appointment or an order.
- Turn an agent’s answer into audio for a messaging channel.
- Let an agent answer with audio, using the component as a tool.
- Produce the audio for a list of texts in one go with Batch Mode.
Configuration
| Field | What it does |
|---|---|
| API Key (required) | Your Deepgram key. The key is filled in for you when you choose your Deepgram credential. |
| Text to Speech (required) | The text to read out. You can type it or connect it from another component. |
| Voice | The Aura-2 voice. Default: aura-2-celeste-es. The voice sets the language: those ending in -es read Spanish (there are 17) and those ending in -en read English. You can also type the name of another voice from Deepgram’s list. |
Advanced options
| Field | What it does |
|---|---|
| Speed | The reading speed, from 0.7 to 1.5. Default: 1, the voice’s natural pace. Spanish voices sound best at 0.9 or above. |
| Audio format | The file format: mp3 (default), wav or ogg. MP3 is small and plays everywhere (WhatsApp, email, browser); WAV is uncompressed; OGG suits voice notes. |
Output
The Data output returns an audio file with the text read out, in the chosen format. Its name is the first words of the text, without accents or punctuation (for example, Hola_buenos_dias_Su.mp3).
Connect it to a component that stores or sends files. An agent can also use the component as a tool to create audio when it needs to; see what tools are in Key concepts.
Example
An online shop sends a WhatsApp notice when an order leaves the warehouse. The flow writes the notice in the customer’s language and picks the voice to match: aura-2-celeste-es for Spanish-speaking customers and aura-2-thalia-en for English-speaking ones. The audio is sent along with the tracking number.
Related components
- Speech to text and text to speech: every component that transcribes audio or produces it.
- Deepgram Speech to Text: the opposite, turns audio into text with the same credential.
- Supertonic TTS: voices that work with no credential and no internet.
Tips
- Check that the voice and the text are in the same language: an
-esvoice reading an English text sounds wrong. - Measured in our test environment on 3 November 2026: the MP3 of the sentence “Hola, buenos días. Su cita está confirmada para el martes a las diez y media.” was produced in 1.1 s.