Skip to navigation
Get started

TTS quickstart

Make your first TTS request and save the result as a WAV file.
View as Markdown

In this quickstart, you will send one HTTPS request, save the generated audio as speech.wav, and play it locally.

Low time-to-first-audio (TTFA) requires stream: true. This quickstart covers basic usage and waits for the complete audio to be generated. For faster playback startup, follow the streaming audio guide to receive and play audio as it arrives.

1. Create an API key

Create a key in API Keys and copy it when it is shown. Store it in your server environment:

export NARI_API_KEY="your-api-key"

Keys are server-side credentials. The authentication guide covers key management and browser applications.

2. Generate speech

voice is required and case-sensitive. Use diana for Diana (female) or leon for Leon (male), rather than the display names Diana or Leon. The voice catalog lists the IDs available for each model.

curl --fail --silent --show-error \
  https://api.narilabs.com/v1/audio/speech \
  -H "Authorization: Bearer $NARI_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "qwen3-tts",
    "voice": "diana",
    "input": "Hello, welcome to Nari Labs.",
    "response_format": "wav"
  }' \
  --output speech.wav

3. Play the result

Open speech.wav in your audio player. On macOS, run:

afplay speech.wav

If you hear “Hello, welcome to Nari Labs,” your request succeeded.

For best results, write input with normal capitalization and punctuation.

Why this model

Qwen3-TTS is best suited to output-streaming applications where low time-to-first-audio (TTFA) matters. This quickstart uses a WAV file to keep the first request simple; the streaming audio guide shows how to play audio as it arrives and explains the streaming response format.

For applications that receive text incrementally, use the TTS input streaming guide.

Next steps