On a slightly different tack let's go back to the basic derivation of Doppler shift. Suppose the source is moving towards to observer. Then the wavelength is shortened by $V_{source} \cdot \Delta T$$V_\textrm{source} \cdot \Delta T$ where $\Delta T$ is the reciprocal of sound frequency. Whereby we derive the shift co-efficient:
$$\frac{V_{sound}}{V_{sound}-V_{source}}$$$$\frac{V_\textrm{sound}}{V_\textrm{sound}-V_\textrm{source}}$$
But what I am struggling to understand is that if rather the observer is moving towards the source, why we may not determine that the wavelength is similarly shortened by $V_{listener} \cdot \Delta T$$V_\textrm{listener} \cdot \Delta T$ and hence derive the shift co-efficient as
$$\frac{V_{sound}}{V_{sound}-V_{listener}}$$$$\frac{V_\textrm{sound}}{V_\textrm{sound}-V_\textrm{listener}}$$
In which case the formula does indeed depend only upon relative motion between source and listener and not whether it is source or observer moving.