Think Sound

Think Sound

Playground

Try it on WaveSpeedAI!

ThinkSound turns uploaded videos into realistic, text-guided audio. Upload a video and add a text prompt to generate lifelike sound. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

ThinkSound is a cutting-edge video-to-audio generation model that creates realistic sound effects and audio tracks from video content. By analyzing the visual elements and actions in your video, it generates high-quality audio that perfectly matches what’s happening on screen.


Why It Stands Out

  • Video-to-audio generation: Converts video content into corresponding audio tracks automatically.
  • Context-aware sound: Analyzes visual elements and actions to generate matching audio.
  • Prompt-guided output: Add text descriptions to guide the style and type of audio generated.
  • Prompt Enhancer: Built-in AI-powered prompt optimization for better audio results.
  • High-quality output: Produces clear, realistic audio that enhances the multimedia experience.
  • Reproducibility: Use the seed parameter to recreate exact results or explore variations.

Parameters

ParameterRequiredDescription
videoYesSource video (upload or public URL).
promptNoText description to guide audio generation style.
seedNoSet for reproducibility; -1 for random.

How to Use

  1. Upload your video — drag and drop a file or paste a public URL.
  2. Add a prompt (optional) — describe the type of sound you want. Use the Prompt Enhancer for AI-assisted optimization.
  3. Set a seed (optional) for reproducible results.
  4. Click Run and wait for audio generation.
  5. Preview and download the result.

Best Use Cases

  • Video Post-Production — Add sound effects to silent footage or animations.
  • Content Creation — Generate audio tracks for social media videos and clips.
  • Game Development — Create sound effects for gameplay footage and trailers.
  • Film & Animation — Produce foley and ambient sounds for scenes.
  • Advertising — Add impactful audio to promotional videos.

Pricing

OutputPrice
Per video$0.05

Pro Tips for Best Quality

  • Use videos with clear visuals and distinct actions for best audio matching.
  • Add prompts to specify the type of sound you want (e.g., “engine roaring”, “footsteps on gravel”).
  • Ensure the video has visible sound-producing elements like movement, impacts, or environmental features.
  • For complex scenes, describe the primary sound you want the model to focus on.
  • Fix the seed when iterating to compare different prompt variations.

Notes

  • Ensure uploaded video URLs are publicly accessible.
  • Processing time varies based on video length and current queue load.
  • Please ensure your content complies with usage guidelines.

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'
{
  "video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
  "seed": -1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/think-sound" \
  -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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 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) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

ParameterTypeRequiredDefaultRangeDescription
videostringYes-The video for generating the output.
promptstringNo-The positive prompt for the generation.
seedintegerNo-1-The random seed to use for the generation. -1 means a random seed will be used.

Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
data.idstringUnique identifier for the prediction, Task Id
data.modelstringModel ID used for the prediction
data.outputsarrayOutput values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed)
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to retrieve the prediction result
data.statusstringStatus of the task: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds

Result Request Parameters

ParameterTypeRequiredDefaultDescription
idstringYes-Task ID

Result Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
dataobjectThe prediction data object containing all details
data.idstringUnique identifier for the prediction
data.modelstringModel ID used for the prediction
data.outputsarray<string | object>Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model.
data.urlsobjectObject containing related API endpoints
data.urls.getstringURL to poll for the prediction result
data.statusstringStatus: created, processing, completed, or failed
data.created_atstringISO timestamp of when the request was created
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds
© 2026 WaveSpeedAI. All rights reserved.