That depends on what kind of clouds they are and the wind speed at their altitude. Clouds do not move on its own. Clouds dissipate in the surrounding air but move in the direction of wind fonly. Therefore speed of clouds cannot be more than that of the wind it is caught up in. In fact the cloud movement is used as a tool to measure the wind at various heights, in Satellite applications, (of meteorology).

A Doppler Wind Lidar emits a laser beam into the atmosphere and detects the reflection from molecules, aerosols and cloud droplets, depending on laser wavelength. The frequency of the reflected light will be slightly shifted due to the movement of the particles, the so-called Doppler shift. The frequency shift is a measure for the particle's mean speed and hence for the ambient wind speed. The time delay between emitted and received laser light allows the retrieval of wind profiles using this technique. ESA's Core Earth Explorer Atmospheric Dynamics Mission (Aeolus) is the first instrument to provide a three-dimensional global coverage of wind observations.

Knowing only the velocity of a cloud (its speed and direction) is, in principle, not enough to know the wind speed close to the surface. Wind speed and direction change with height following the so-called Ekman spiral and the shape of the spiral (i.e., shear of wind speed and the change of wind direction with the height) depends on many factors such as the atmospheric stability, surface roughness, wind speed above the planetary boundary layer, orthography, etc. All these contributors are unknown if you only have the cloud translation velocity.

No it can not be done exactly. The wind itself contribute to the velocity of cloud in time and space.. Wind profile studied at various locations indicated that near the earth surface is low due to the frictional forces and it becomes laminar as we move in the upward direction which hinders the relationship between cloud movement and speed of wind at specific location.

Dear doctor

Seeing the Wind: Visual Wind Speed Prediction with a Coupled Convolutional and Recurrent Neural Network

Jennifer L. Cardona, Michael F. Howland, John O. Dabiri

33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada.

"Abstract Wind energy resource quantification, air pollution monitoring, and weather forecasting all rely on rapid, accurate measurement of local wind conditions. Visual observations of the effects of wind—the swaying of trees and flapping of flags, for example—encode information regarding local wind conditions that can potentially be leveraged for visual anemometry that is inexpensive and ubiquitous. Here, we demonstrate a coupled convolutional neural network and recurrent neural network architecture that extracts the wind speed encoded in visually recorded flow-structure interactions of a flag and tree in naturally occurring wind. Predictions for wind speeds ranging from 0.75-11 m/s showed agreement with measurements from a cup anemometer on site, with a root-mean-squared error approaching the natural wind speed variability due to atmospheric turbulence. Generalizability of the network was demonstrated by successful prediction of wind speed based on recordings of other flags in the field and in a controlled wind tunnel test. Furthermore, physicsbased scaling of the flapping dynamics accurately predicts the dependence of the network performance on the video frame rate and duration."