/**
* Calculates the distance between this voice sample and the voice prints previously extracted
* and returns the closest matches sorted by distance
* <p>
* Usage of a closed set is assumed : the speaker's voice print was extracted before and is known to the system.
* This means you'll always get MatchResults even if the speaker is absolutely unknown to the system.
* The MatchResult class provides a likelihood ratio in order to help determining the usefulness of the result
* </p>
* @param voiceSample the voice sample, values between -1.0 and 1.0
* @return a list MatchResults sorted by distance
*/
public IEnumerable <MatchResult <T> > Identify(double[] voiceSample, double maxDistance = 0)
{
if (store.Count == 0)
{
throw new InvalidOperationException("There is no voice print enrolled in the system yet");
}
var voicePrint = new VoicePrint(audioProcessor.ProcessAndExtract(voiceSample));
var calculator = new EuclideanDistanceCalculator();
var matches = new List <MatchResult <T> >(store.Count);
var distanceFromUniversalModel = voicePrint.GetDistance(calculator, universalModel);
foreach (var entry in store)
{
var distance = entry.Value.GetDistance(calculator, voicePrint);
if (maxDistance > 0 && distance > maxDistance)
{
continue;
}
// likelihood : how close is the given voice sample to the current VoicePrint
// compared to the total distance between the current VoicePrint and the universal model
int likelihood = 100 - (int)(distance / (distance + distanceFromUniversalModel) * 100);
matches.Add(new MatchResult <T>(entry.Key, likelihood, distance));
}
return(matches.OrderBy(m => m.Distance));
}
/**
* Creates a voice print and stores it along with the user key for later comparison with new samples
* <p>
* Threading : this method is synchronized to prevent inadvertently erasing an existing user key
* </p>
* @param userKey the user key associated with this voice print
* @param voiceSample the voice sample, values between -1.0 and 1.0
* @return the voice print extracted from the given sample
*/
public VoicePrint CreateVoicePrint(T userKey, double[] voiceSample)
{
lock (_lock)
{
if (userKey == null)
{
throw new ArgumentNullException(nameof(userKey), "The userKey is null");
}
if (store.ContainsKey(userKey))
{
throw new ArgumentException("The userKey already exists: [{userKey}");
}
double[] features = audioProcessor.ProcessAndExtract(voiceSample);
VoicePrint voicePrint = new VoicePrint(features);
if (!universalModelWasSetByUser)
{
if (universalModel == null)
{
universalModel = new VoicePrint(voicePrint);
}
else
{
universalModel.Merge(features);
}
}
store.Add(userKey, voicePrint);
return(voicePrint);
}
}
/**
* Sets the universal model to be used to calculate likelihood ratios
* Once set, further voice print create / merge operations won't modify this model
* @param universalModel the universal model to set, may not be null
*/
public void SetUniversalModel(VoicePrint universalModel)
{
lock (_lock)
{
this.universalModel = universalModel ?? throw new ArgumentNullException(nameof(universalModel), "The universal model may not be null");
universalModelWasSetByUser = false;
}
}
/**
* Returns the distance between this voice print and the given one using the calculator.
* Threading : it is safe to call this method while other threads may merge this voice print instance
* with another one in the sense that the distance calculation will not happen on half merged voice print.
* Since this method is read only, it is safe to call it from multiple threads for a single instance
* @param calculator the distance calculator
* @param voicePrint the voice print
* @return the distance
*/
public double GetDistance(DistanceCalculator calculator, VoicePrint voicePrint)
{
rwl.EnterReadLock();
try
{
return(calculator.GetDistance(features, voicePrint.features));
}
finally
{
rwl.ExitReadLock();
}
}
/**
* Constructor taking previously extracted voice prints directly into the system
* @param sampleRate the sample rate, at least 8000.0 Hz (preferably higher)
* @param voicePrintsByUserKey a {@code Map} containing user keys and their respective {@code VoicePrint}
*/
public Recognito(float sampleRate, Dictionary <T, VoicePrint> voicePrintsByUserKey) : this(sampleRate)
{
VoicePrint universalModel = null;
foreach (var item in voicePrintsByUserKey.Values)
{
if (universalModel == null)
{
universalModel = new VoicePrint(item);
continue;
}
universalModel.Merge(item);
}
store.AddRange(voicePrintsByUserKey);
}
/**
* Copy constructor
* @param print the VoicePrint to copy
*/
public VoicePrint(VoicePrint print) : this(ArrayHelper.Copy(print.features, print.features.Length))
{
}
/**
* Convenience method to merge voice prints
* @param print the voice print to merge
* @see VoicePrint#merge(double[])
*/
public void Merge(VoicePrint print)
{
Merge(print.features);
}
