Variadic template class: Allow for float and int arguments cast to float?

I have the following class:

template <size_t N, float... Lam> requires (sizeof...(Lam) == N)
class Spectrum {
public:
    Spectrum() : wavelengths{ static_cast<float>(Lam)... }
    {
        values.fill(0.f);
    };

    template <IsNumeric T>
    Spectrum(T v) : wavelengths{ static_cast<float>(Lam)... }
    {
        values.fill(static_cast<float>(v));
    };

    template <IsNumeric... T> requires (sizeof...(T) == N)
    Spectrum(T... v) : wavelengths{ static_cast<float>(Lam)... }, values { static_cast<float>(v)... } 
    { };

... // Other methods/overloads that aren't relevant here

private:
    std::array<float, N> values;
    const std::array<float, N> wavelengths;
};

I can then call this using:

Spectrum<3, 450.f, 500.f, 600.f> color{ 1,2,3 };

And it behaves as expected (the wavelengths are populated with 450.f, 500.f, 600.f and the values are populated with 1.f, 2.f, 3.f.

I would like to be able to allow users to instead call:

Spectrum<3, 450, 500, 600> color{1,2,3};

in this case, 450 is an int, so I'd need it to be cast to a float, in a similar way to what I already do in the constructor templates. I'm just not sure if a similar idea can be expressed in the class template itself.

template<auto...Lam>
using Spectrum_t = Spectrum<sizeof...(Lam), static_cast<float>(Lam)...>;

they don't even have to pass 3. Not sure why you are passing it.

Also, storing Lam... in const wavelengths seems like a very strange choice.

static constexpr std::array<float, N> wavelengths() {
  return {{Lam...}};
}

or an inline constexpr static value.