Lipsyncfrom$0.1000 / s

Wan 2.2 Speech to Video

Makes a face say an audio track · 480p to 720p

image-to-lipsync

curl -X POST https://api.open.sume.com/v1/video-router/generate \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-2.2-speech-to-video","image_url":"https://example.com/image.png","audio_url":"https://example.com/audio.mp3","resolution":"480p"}'

1. Calling the API

Wan 2.2 Speech to Video has the model id wan-2.2-speech-to-video. Send a JSON body to POST https://api.open.sume.com/v1/video-router/generate.

Parameter	Type	Required	Default	Description
model	string	Yes	—	"wan-2.2-speech-to-video"
image_url	string	Yes	—	The face that speaks (HTTPS image).
audio_url	string	Yes	—	The speech (HTTPS audio).
resolution	string	No	480p	Output resolution. Options: 480p, 580p, 720p.

2. Authentication

Send your open.sume.com key as a bearer token from server code only. Keys minted on open.sume.com bill this balance in dollars and call the model API only.

Authorization: Bearer $SUME_API_KEY

3. Queue

A submit answers 202 with the job, its status_url and result_url. Poll the status until it is completed, failed or canceled; failed and canceled jobs are refunded.

GET https://api.open.sume.com/v1/jobs/<job_id>/status
GET https://api.open.sume.com/v1/jobs/<job_id>/result
POST https://api.open.sume.com/v1/jobs/<job_id>/cancel

Pricing

Dollars per unit at each option, from our rate card; a request rounds up to $0.0001. The estimate above is the exact charge.

Option	Price
480p	$0.1000 / s
580p	$0.1500 / s
720p	$0.2000 / s

More models by Alibaba

Price, resolution and duration show each model's full range.