What is the difference between verification and validation?

1. Diagram
   Observation

   Handwritten text 'E = Elastic Modulus' appears on the left side of the screen.

2. Audio
   Observation

   The speaker says, 'the Young's modulus of the beam is E'.

E

Elastic Modulus (Young's modulus) of the beam material

Material property

1. Diagram
   Observation

   Handwritten text 'I = Second moment of area' appears below the definition of E.

2. Formula
   Observation

   The equation $I = \iint z^2 dy dz$ is written on the screen.

I

Second moment of area of the beam cross-section

Geometric property

1. Diagram
   Observation

   Handwritten text '$\mu$ = mass per unit length' appears below the definition of I.

2. Audio
   Observation

   The speaker says, 'and $\mu$ is the mass per unit'.

$\mu$

Mass per unit length of the beam

Physical property

1. Diagram
   Observation

   A red arrow pointing downwards at the tip of the beam is labeled 'F'.

2. Audio
   Observation

   The speaker says, 'with a value of F'.

F

Applied force at the tip of the cantilever beam

Load

1. Diagram
   Observation

   A blue vertical dimension line indicating deflection is labeled 'w'.

2. Audio
   Observation

   The speaker says, 'and the deflection of beam is w'.

w

Deflection of the beam at the point of force application

Displacement

1. Diagram
   Observation

   In the Verification box, a distributed load is shown with red arrows labeled '$q(x)$'.

2. Formula
   Observation

   The term $q(x)$ appears on the right-hand side of the PDE.

$q(x)$

Distributed transverse load as a function of position x

Load distribution

1. Diagram
   Observation

   An axis along the length of the beam in the Verification box is labeled 'x'.

2. Formula
   Observation

   The variable x appears in the partial derivative operators $\frac{\partial}{\partial x}$.

x

Spatial coordinate along the length of the beam

Position

1. Formula
   Observation

   The handwritten PDE on the upper diagram contains $q(x)$ on the right-hand side.

2. Audio
   Observation

   The speaker says, "Loading is QX for a general form".

$q(x)$

General distributed loading function acting on the beam.

Function of spatial coordinate x

1. Formula
   Observation

   The handwritten PDE contains w in the terms $\frac{\partial^2 w}{\partial x^2}$ and $\frac{\partial^2 w}{\partial t^2}$.

2. Audio
   Observation

   The speaker says, "when the deflection is W as a function of X".

w

Deflection of the beam.

Function of x (and potentially t in the general equation)

1. Formula
   Observation

   The handwritten PDE contains x in the denominator of the partial derivative $\frac{\partial^2}{\partial x^2}$.

2. Audio
   Observation

   The speaker says, "as a function of X".

x

Spatial coordinate along the length of the beam.

Real number

1. Formula
   Observation

   The handwritten PDE contains EI inside the parentheses.

2. Audio
   Observation

   The speaker says, "So we have a Young modulus and we have also Q in this equation".

1. The speaker mentions 'Young modulus' but the formula uses the product EI (flexural rigidity). The symbol I is not explicitly named in the audio.

EI

Flexural rigidity of the beam, where E is Young's modulus and I is the area moment of inertia.

Positive real constant

1. Formula
   Observation

   The handwritten PDE contains $\mu$ multiplying the time derivative term.

1. The symbol $\mu$ is not mentioned in the audio.

$\mu$

Mass per unit length of the beam.

Positive real constant

1. Audio
   Observation

   Speaker states: 'verification is simply when we evaluate our model's prediction based on the mathematics of the model... make sure that the solution... is mathematically correct.'

2. Diagram
   Observation

   Text 'Verification -> mathematics' is displayed on the slide.

Verification is the process of evaluating a numerical model's predictions based strictly on the underlying mathematical equations to ensure the solution is mathematically correct.

1. Applies to numerical simulations

2. Focuses on mathematical correctness

1. Audio
   Observation

   Speaker states: 'in the validation, we want to make sure that it follows the physical assumptions and the physical meaning... almost the same as our experimental measurement.'

2. Diagram
   Observation

   Text 'Validation -> Physics' is displayed on the slide.

Validation is the process of ensuring that the model's predictions align with physical assumptions and match experimental measurements closely.

1. Applies to numerical simulations

2. Focuses on physical realism and experimental agreement

1. Formula
   Observation

   Equation $\frac{\partial^2}{\partial x^2}$ (EI $\frac{\partial^2 w}{\partial x^2}$) + $\mu \frac{\partial^2 w}{\partial t^2} = q(x)$ is written in the Verification box.

2. Audio
   Observation

   Speaker mentions 'fourth order equation of motion... it's a PDE'.

1. The time variable t is not explicitly defined in the audio but appears in the formula.

A fourth-order partial differential equation describing the dynamic behavior of a beam under distributed load, involving bending stiffness and inertia.

1. Applicable to Euler-Bernoulli beams

2. Assumes small deflections

1. E
2. I
3. $\mu$
4. $q(x)$
5. x

1. Formula
   Observation

   The handwritten equation is $\frac{\partial^2}{\partial x^2} \left( EI \frac{\partial^2 w}{\partial x^2} \right) + \mu \frac{\partial^2 w}{\partial t^2} = q(x)$.

2. Audio
   Observation

   The speaker explains the terms: "Loading is QX... deflection is W... Young modulus...".

The video presents the general partial differential equation governing the dynamic deflection of a beam under distributed loading.

1. Applies to a beam with flexural rigidity EI and mass per unit length $\mu$.

1. Audio
   Observation

   The speaker states, "But we can neglect the acceleration term, so we can consider as a static deflection."

To simplify the problem for numerical verification, the time-dependent acceleration term is neglected, reducing the dynamic PDE to a static boundary value problem.

1. The loading and deflection are assumed to be independent of time.

1. General Beam Deflection PDE

1. Formula
   Observation

   The handwritten equation is $w = \frac{FL^3}{3EI}$.

2. Audio
   Observation

   The speaker says, "the analytical solution is W equal FL to the power of 3 over 3EI".

For a cantilever beam with a point load at the free end, the exact deflection at the tip is given by this standard structural mechanics formula.

1. Cantilever boundary condition (clamped at one end, free at the other).

2. Point force F applied vertically at the free tip.

3. Linear elastic material behavior.

1. Static Deflection Assumption

1. Formula
   Observation

   The handwritten text lists "FEM" and "FDM" under "Numerical solution".

2. Audio
   Observation

   The speaker says, "for instance we want to use finite difference method or finite element method".

The Finite Element Method (FEM) and Finite Difference Method (FDM) are numerical techniques used to approximate the solution to the beam deflection differential equation when an analytical solution is unavailable or to verify code implementations.

1. General Beam Deflection PDE

1. Audio
   Observation

   The speaker explains, "we reduce the mesh element, the element size... by increasing the number of element it converges to a certain value".

2. Diagram
   Observation

   A hand-drawn graph shows a curve approaching a horizontal dashed line as the x-axis value increases.

In numerical methods like FEM, the domain is divided into elements. Reducing the element size increases the degrees of freedom (DOF). A key verification step is to observe that the computed deflection converges to the true analytical value as the DOF increases.

1. Requires a numerical solver (e.g., FEM or FDM).

2. Assumes the analytical solution is known for comparison.

1. Numerical Methods for Beam Deflection
2. Analytical Solution for Cantilever Tip Deflection

1. Audio
   Observation

   Speaker says 'we can say that it's verified'.

2. Diagram
   Observation

   Graph shows numerical solution converging to analytical solution w = FL^$3 / 3$EI.

Verification ensures that the numerical solution (e.g., FEM or FDM) converges to the analytical solution as the mesh is refined (DOF increases).

1. Analytical solution exists

2. Mesh refinement

1. Audio
   Observation

   Speaker says 'For validation, we make sure that our solution is equal to the measurement'.

2. Diagram
   Observation

   Drawing of an experimental setup with an I-beam, welding, and a mass m.

Validation compares the simulation results with physical experimental measurements to ensure the model accurately represents reality.

1. Experimental setup available

2. Measurement data obtained

1. Audio
   Observation

   Speaker says 'we can compute the error between these two... In fact, it's not error, it's relative difference'.

2. Formula
   Observation

   Written as 'Rel dif %' on the whiteboard.

A metric used in validation to quantify the discrepancy between experimental and simulation results, expressed as a percentage.

1. Both experimental and simulation values are available

1. Validation

1. Audio
   Observation

   Speaker says 'our validation metric is the displacement at the tip of the beam... Or we could also define different validation metric... measure the epsilon E11'.

2. Diagram
   Observation

   Drawing of a beam with a strain gauge and deflection w.

The specific physical quantity chosen to compare between the experiment and the simulation, such as tip displacement or strain.

1. Clearly defined physical quantity

1. Validation

1. Audio
   Observation

   The speaker states, "by increasing the number of element it converges to a certain value and this certain value should be equal to W FL3".

2. Diagram
   Observation

   The drawn curve asymptotically approaches the dashed line labeled with the analytical solution.

As the number of elements (or degrees of freedom) in a numerical model increases, the computed deflection converges to the exact analytical solution.

1. The numerical method (e.g., FEM) is correctly implemented.

2. The problem has a known analytical solution for comparison.

For all sufficiently fine meshes.

1. Audio
   Observation

   Speaker says 'it depends on our application what we want to define our validation metric'.

The choice of validation metric depends on the specific application requirements.

No quantifiers are stated in the source.

1. Audio
   Observation

   Speaker says 'when we validate and verify our simulation then we can use it in our application'.

A simulation can be used in its intended application only after it has been both verified and validated.

No quantifiers are stated in the source.

1. Audio
   Observation

   The speaker explains the transition from the general PDE to the static case by neglecting the acceleration term.

1. Expression

   $$$$
   Explanation

   Start with the general dynamic PDE: $\frac{\partial^2}{\partial x^2} \left( EI \frac{\partial^2 w}{\partial x^2} \right) + \mu \frac{\partial^2 w}{\partial t^2} = q(x)$.

   Justification

   Observed from the whiteboard.

   Shown in the video
2. Expression

   $$$$
   Explanation

   Neglect the acceleration term $\mu \frac{\partial^2 w}{\partial t^2}$.

   Justification

   Stated by the speaker to consider static deflection.

   Shown in the video
3. Expression

   $$$$
   Explanation

   This leaves the static beam equation, which is then solved numerically for the specific cantilever case.

   Justification

   Logical consequence of the previous step.

   Derived from the video

The problem is simplified to a static boundary value problem suitable for numerical verification against a known analytical solution.

1. Audio
   Observation

   Speaker describes measuring experimental strain and computing simulated strain at integration points to compare them.

2. Diagram
   Observation

   Drawing of a beam with '$\epsilon_{11}$' and a cross-section with nodes and '$\varepsilon$'.

1. Expression

   $$\epsilon_{11} \text{ (experimental)}$$
   Explanation

   Measure strain at a specific location on the physical beam.

   Justification

   Audio: 'I could measure epsilon one one'.

   Shown in the video
2. Expression

   $$\varepsilon \text{ (simulated)}$$
   Explanation

   Compute strain at the corresponding integration point in the FEA model.

   Justification

   Audio: 'compute the epsilon at the integration point'.

   Shown in the video
3. Expression

   $$\text{Compare } \epsilon_{11} \text{ and } \varepsilon$$
   Explanation

   Use the difference as a validation metric.

   Justification

   Audio: 'compare these two epsilon as a different validation metric'.

   Shown in the video

Comparing experimental and simulated strains provides a valid metric for model validation.

1. Diagram
   Observation

   Drawing of a cantilever beam fixed at one end with force F at the other.

2. Audio
   Observation

   Speaker describes the problem setup: 'imagine that this is our problem... beam when the force is applied at the tip...'

Determine the deflection w of a cantilever beam subjected to a tip force F.

1. Beam fixed at one end (cantilever)

2. Force F applied at the free tip

3. Material property E (Elastic Modulus)

4. Geometric property I (Second moment of area)

5. Mass per unit length $\mu$

Calculate deflection w using analytical and numerical methods for verification, and compare with experiment for validation.

1. Expression

   $$w = \frac{FL^3}{3EI}$$
   Explanation

   Analytical solution for static deflection of a cantilever beam with tip load.

   Justification

   Standard beam theory result shown in the Verification box.

   Shown in the video
2. Expression

   $$FEM / FDM$$
   Explanation

   Numerical solutions obtained via Finite Element Method or Finite Difference Method.

   Justification

   Listed as numerical approaches in the Verification box.

   Shown in the video

The analytical deflection is w = FL^3/(3EI). Numerical methods like FEM and FDM are used to approximate this solution for verification.

Comparison of numerical results with the analytical solution (Verification) and comparison with experimental Force-Displacement data (Validation).

1. Diagram
   Observation

   A drawing of a cantilever beam with a point load F at the tip and deflection w.

2. Formula
   Observation

   The analytical solution $w = \frac{FL^3}{3EI}$ is written below the diagram.

3. Audio
   Observation

   The speaker describes the setup: "clamped boundary condition... vertical and pointed force like F... compute the deflection at the tip".

Verify a numerical implementation (e.g., FEM) for solving beam deflection.

1. A cantilever beam of length L and flexural rigidity EI.

2. A vertical point force F applied at the free tip.

3. The known analytical solution for tip deflection: $w = \frac{FL^3}{3EI}$.

Demonstrate that the numerical solution converges to the analytical solution as the mesh is refined.

1. Expression

   $$$$
   Explanation

   Build the FEM model and write the code.

   Justification

   Stated by the speaker as the first step.

   Shown in the video
2. Expression

   $$$$
   Explanation

   Run the simulation with progressively smaller element sizes (increasing DOF).

   Justification

   Stated by the speaker: "reduce the mesh element, the element size".

   Shown in the video
3. Expression

   $$$$
   Explanation

   Plot the computed tip deflection w against the number of DOF.

   Justification

   Shown in the hand-drawn graph.

   Shown in the video
4. Expression

   $$$$
   Explanation

   Observe that the plotted curve converges to the horizontal line representing the analytical value $\frac{FL^3}{3EI}$.

   Justification

   Stated by the speaker and shown in the graph.

   Shown in the video

The numerical method is verified if the computed deflection approaches the analytical solution $\frac{FL^3}{3EI}$ as the degrees of freedom increase.

Comparison of the numerical result's limit with the exact analytical formula.

1. Diagram
   Observation

   Drawing of an I-beam welded to a steel plate with a mass m at the tip.

2. Audio
   Observation

   Speaker describes the setup: 'we have an I-beam here... welding... join the beam over a plate... apply a force by adding a mass m at the tip'.

Describe the experimental setup used for validating the beam deflection simulation.

1. I-beam made of steel ($E=200$ GPa, $ν=0.3$)

2. Steel plate ($E=800$ GPa, $ν=0.3$)

3. Welded joint

4. Mass m at the tip

Identify components and boundary conditions of the physical experiment.

1. Expression

   $$$$
   Explanation

   An I-beam is welded to a steel plate.

   Justification

   Visual observation of the diagram.

   Shown in the video
2. Expression

   $$$$
   Explanation

   A mass m is hung at the free tip of the I-beam to apply a downward force.

   Justification

   Visual observation and audio description.

   Shown in the video
3. Expression

   $$$$
   Explanation

   The boundary condition at the weld is not perfectly clamped, allowing some deflection.

   Justification

   Audio statement: 'it's not completely clamped boundary condition so we may have a deflection'.

   Shown in the video

The setup consists of a steel I-beam welded to a stiffer steel plate, with a mass applied at the tip to measure force-displacement behavior.

Matches the visual diagram and audio description.

1. Formula
   Observation

   Visible formulas: '$w = \frac{FL^3}{3EI}$', '$\frac{\partial^2}{\partial x^2}(EI \frac{\partial^2 w}{\partial x^2}) + \mu \frac{\partial^2 w}{\partial t^2} = q(x)$'.

2. Diagram
   Observation

   Box labeled 'Verification' containing analytical and numerical solutions for a beam.

Verify the numerical solution (FEM/FDM) against the analytical solution for beam deflection.

1. Analytical solution: $w = \frac{FL^3}{3EI}$

2. Numerical solution: FEM, FDM

Demonstrate that the numerical method converges to the analytical result.

1. Expression

   $$w_{analytical} = \frac{FL^3}{3EI}$$
   Explanation

   State the exact analytical solution for deflection.

   Justification

   Visible in the 'Verification' box.

   Shown in the video
2. Expression

   $$w_{numerical} \in \{\text{FEM}, \text{FDM}\}$$
   Explanation

   Identify numerical methods used to approximate the solution.

   Justification

   Visible in the 'Verification' box.

   Shown in the video

Verification ensures the numerical solver correctly solves the governing differential equation.

Comparison of $w_{numerical}$ and $w_{analytical}$ on the graph.

1. Diagram
   Observation

   Box labeled 'Validation' showing a physical steel beam setup and a Force-Displacement graph comparing 'Experiment' and 'Simulation'.

Validate the simulation model against physical experimental data for a steel beam.

1. Material: Steel ($E=200 \text{ GPa}$, $\nu=0.3$)

2. Setup: Welded I-beam with mass $m$

Ensure the simulation accurately predicts real-world behavior.

1. Expression

   $$\text{Force [N]} \text{ vs } \text{Displacement [mm]}$$
   Explanation

   Plot experimental and simulation results on the same axes.

   Justification

   Visible in the 'Validation' box graph.

   Shown in the video
2. Expression

   $$\text{Experiment line} \approx \text{Simulation line}$$
   Explanation

   Compare the two curves to assess model accuracy.

   Justification

   Visual representation in the graph.

   Shown in the video

Validation confirms the model represents the physical reality of the steel beam structure.

Closeness of the blue 'Experiment' line and green 'Simulation' line on the graph.

1. Animation
   Observation

   The view zooms out from the title slide to reveal handwritten notes and diagrams on a digital whiteboard.

1. Title Slide

2. Handwritten Notes

3. Diagrams

1. Zoom level decreases

2. Content shifts from typed text to handwritten math

1. Topic remains Verification and Validation

The instructor moves from conceptual definitions to a concrete mathematical example.

1. Animation
   Observation

   Two large rectangular boxes are drawn around the 'Verification' and 'Validation' sections respectively.

1. Verification Box

2. Validation Box

1. Visual grouping of related equations and diagrams

1. Underlying math content unchanged

Emphasizes the distinct workflows for checking math (Verification) vs checking physics (Validation).

1. Diagram
   Observation

   Top section of the whiteboard showing a beam with distributed load $q(x)$, deflection $w(x)$, and the general PDE.

1. Cantilever beam

2. Distributed load arrows (red)

3. Deflection curve (blue dashed)

4. PDE equation (black)

1. The beam is fixed at the left end.

Illustrates the general dynamic loading scenario before simplification to the static case.

1. Diagram
   Observation

   Middle section of the whiteboard showing a cantilever with point load F, deflection w, and the analytical solution formula.

1. Cantilever beam

2. Point load arrow (red)

3. Deflection curve (blue dashed)

4. Analytical solution formula (black)

5. List of numerical methods (FEM, FDM)

1. The beam is fixed at the left end and free at the right.

Defines the simple benchmark problem used to test the numerical code.

1. Animation
   Observation

   A coordinate system is drawn, followed by axes labels (DOF, w), a horizontal dashed line, and a curve with points approaching the line.

1. Horizontal axis (DOF)

2. Vertical axis (w)

3. Horizontal dashed line (analytical solution)

4. Curve with discrete points (numerical results)

1. Axes are drawn.

2. Labels DOF and w are added.

3. A horizontal dashed line is drawn representing the target value.

4. A curve with points is drawn, starting low and rising to approach the dashed line.

1. The horizontal dashed line remains at a constant height.

Visually demonstrates the concept of mesh convergence: as the numerical model becomes more complex (higher DOF), its output gets closer to the true value.

1. Diagram
   Observation

   A graph plotting w vs DOF, showing a curve approaching a horizontal dashed line representing the analytical solution.

1. w-axis

2. DOF-axis

3. Curve

4. Dashed line

1. Curve approaches the dashed line as DOF increases.

1. Dashed line remains constant at w = FL^$3 / 3$EI.

Illustrates the verification process where numerical solutions converge to the analytical solution with mesh refinement.

1. Diagram
   Observation

   Hand-drawn sketch of an I-beam welded to a vertical plate, with a mass hanging from the end.

1. I-beam

2. Steel plate

3. Weld

4. Mass m

1. None, static drawing.

1. Material properties E and ν are written next to the beam and plate.

Provides the physical context for the validation step, showing how the force is applied and measured.

1. Diagram
   Observation

   A graph plotting Force [N] vs Displacement [mm] with two lines: blue for Experiment and green for Simulation.

1. Force-axis

2. Displacement-axis

3. Blue line

4. Green line

1. Lines are drawn to show comparison.

1. Both lines start from the origin.

Visualizes the comparison between experimental measurements and simulation predictions for validation.

1. Animation
   Observation

   User draws a rectangular cross-section, adds nodes, and writes '$\varepsilon$'.

1. Rectangular cross-section

2. Nodes

3. Strain symbol $\varepsilon$

1. Drawing lines to form a rectangle

2. Adding dots for nodes

3. Writing epsilon symbol

1. Existing beam diagram above

Illustrates where strain is computed in a finite element mesh (at integration points/nodes).

1. Animation
   Observation

   Canvas zooms out to reveal 'Verification' and 'Validation' boxes side-by-side.

1. Verification box

2. Validation box

3. Equations

4. Graphs

1. Scale decreases

2. More content becomes visible

1. Relative positions of boxes

Provides a holistic view comparing the mathematical verification process with the physical validation process.

1. Audio
   Observation

   The speaker explicitly contrasts the general equation with the static assumption: "But we can neglect the acceleration term, so we can consider as a static deflection."

One might assume the full dynamic PDE must always be solved.

For verification purposes involving static loads, the time-dependent acceleration term can be safely neglected to simplify the problem to a static boundary value problem.

1. Audio
   Observation

   Speaker explicitly corrects himself: 'In fact, it's not error, it's relative difference'.

Confusing absolute error with relative difference when comparing simulation and experiment.

Relative difference is a normalized percentage metric, making it more suitable for comparing results across different scales than absolute error.

1. Diagram
   Observation

   Slide text contrasts 'Verification -> mathematics' with 'Validation -> Physics'.

2. Audio
   Observation

   Speaker explicitly defines both terms in contrast to each other.

Verification checks mathematical correctness against model equations, while Validation checks physical realism against experimental data.

1. Formula
   Observation

   The general PDE is shown above the specific analytical solution w=FL^$3/3$EI.

1. The derivation steps from PDE to analytical solution are skipped in the video.

The general equation of motion applies to the specific cantilever beam example, though the video jumps directly to the static analytical solution.

1. Audio
   Observation

   The speaker transitions from discussing the general PDE to the static assumption.

The static deflection assumption is a special case of the general dynamic PDE where the acceleration term is zero.

1. Audio
   Observation

   The speaker introduces the analytical solution immediately after defining the static problem.

The analytical solution formula applies specifically to the static cantilever beam problem defined by the simplified equation.

1. Audio
   Observation

   The speaker links the use of numerical methods to the process of checking convergence.

Mesh refinement is a core procedure within the application of numerical methods like FEM to ensure accuracy.

1. Audio
   Observation

   The speaker states the converged value should equal the analytical solution.

The verification process relies on comparing the limit of the mesh refinement sequence against the known analytical solution.

1. Audio
   Observation

   Speaker transitions from discussing verification (analytical vs numerical) to validation (simulation vs experiment).

Verification checks the mathematical correctness of the numerical method against an analytical solution, while validation checks the physical accuracy of the model against experimental data.

1. Audio
   Observation

   Speaker defines w and ε11 as examples of validation metrics.

The validation process requires selecting a specific metric, such as tip displacement or strain, to quantify the comparison.

1. Audio
   Observation

   Speaker links verification and validation as prerequisites for using the simulation.

2. Diagram
   Observation

   Arrow drawn from Validation box to a circle representing the application.

Verification and Validation must be completed before the simulation is trusted for application.

1. Audio
   Observation

   Speaker introduces the session by stating he will introduce two concepts: verification and validation.

1. Definition of Verification
2. Definition of Validation

1. Formula
   Observation

   The fourth-order PDE for beam motion is written on the board.

1. Equation of Motion for a Beam

1. Audio
   Observation

   The entire segment discusses using a known solution to check a numerical method.

1. Numerical Methods for Beam Deflection
2. Mesh Refinement and Convergence
3. Analytical Solution for Cantilever Tip Deflection

1. Audio
   Observation

   The speaker explains dropping the acceleration term.

1. General Beam Deflection PDE
2. Static Deflection Assumption

1. Formula
   Observation

   The formula w = FL^$3 / 3$EI is written on the board.

1. Analytical Solution for Cantilever Tip Deflection

1. Audio
   Observation

   Title of the video is 'What is the difference between verification and validation?'.

1. Verification
2. Validation

1. Audio
   Observation

   Speaker discusses comparing epsilons as a validation metric.

1. Strain-based Validation Metric

1. Audio
   Observation

   Speaker distinguishes between the verification and validation boxes.

1. Verification vs. Validation