Google Gemini 3.8 Flash Lite Text To Speech API Documentation

Google Gemini 3.8 Flash Lite Text To Speech API Documentation

Playground

Gemini 3.8 Flash-Lite Text-to-Speech generates expressive speech from text with configurable voice and delivery-style controls for narration, dialogue, localization, virtual assistants, and scalable voice production workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Gemini 3.8 Flash-Lite Text-to-Speech generates expressive WAV speech from text with 30 preset voices and natural-language control over tone, pacing, accent, and emotion.

It supports both single-speaker narration and two-speaker dialogue, making it suitable for voiceovers, conversations, storytelling, and other speech-generation workflows.

Why Choose This?

• Expressive text-to-speech
  Generate natural speech with control over tone, pace, accent, and emotion.

• 30 preset voices
  Choose from a wide range of built-in voices for different speaking styles.

• Natural-language delivery control
  Use style_instructions to describe how the speech should sound.

• Single-speaker narration
  Create voiceovers, narration, explanations, and spoken content.

• WAV output
  Receive generated speech as a downloadable WAV audio file.

Parameters

How to Use

1. Enter the text — Provide the content you want spoken.
2. Choose a voice — Select one of the available preset voices.
3. Add style instructions optional — Describe the desired tone, pace, accent, or emotion.
4. Submit — Generate and retrieve the WAV audio.

For dialogue, configure two speakers and provide their lines through turns.

Pricing

Pricing is $0.04 per 1,000 billable characters, with a minimum charge of $0.04 per request.

Text and delivery instructions contribute to billable character count.

Best Use Cases

• Voiceovers — Generate speech for videos, presentations, and promotional content.
• Narration — Create spoken content for stories, explainers, and educational material.
• Two-speaker conversations — Generate dialogue between two different preset voices.
• Character voices — Use voice and style controls for different speaking personalities.
• Localized content — Guide accent, tone, and delivery for different language workflows.
• Conversational prototypes — Create spoken interactions for demos, assistants, and characters.

Pro Tips

• Keep spoken content separate from delivery instructions.
• Use short, specific style instructions for more consistent results.
• Describe tone, pace, energy, or emotion directly.
• Use different voices for clearer separation in dialogue.
• Inline vocal events such as <laugh> or <sigh> can be used when needed.

Notes

• text supports up to 8000 characters.
• style_instructions supports up to 2000 characters.
• The default voice is Kore.
• Dialogue mode supports two speakers.

Related Models

• Gemini 3.8 Flash-Lite Text-to-Speech — Generate expressive speech with preset voices and lightweight text-to-speech workflows.
• Gemini 3.8 Flash Text-to-Speech — Generate expressive single-speaker or two-speaker speech with natural-language delivery control.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "voice": "Kore"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/google/gemini-3.8-flash-lite/text-to-speech" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters