Generating, Editing, Speaking, and Persisting Pronunciations

The LexiconKit management class is designed to provide a lot of flexibility and minimize the programming necessary to manage lexicon word pronunciations.

Pronunciations and Phonemes

Lexicon word pronunciations are comprised of phonemes or basic units of sounds. Phonemes collectively are represented in an alphabet format. Speech engine vendors have unique phoneme and pronunciation formats (i.e., alphabets) and may support International Phonetic Alphabet (IPA).

For example, the following table illustrates the differences in default pronunciations for the word tomato across speech engines.

IPA	Cepstral Swift	Microsoft SAPI 5	Microsoft Universal Phone Set (UPS)
təme͡ito	t ah0 m ey1 t ow0	t ax m ey t ow	T AX M EI T O

LexiconKit handles the complexities of dealing with these differences for applications.

Generating and Speaking Pronuncations

To generate a lexicon word pronunciation, simply pass the word, the word type (i.e., part of speech), language, and alphabet to LexiconKit. To speak a lexicon word pronunciation, simply pass the phonemes, language, and alphabet to LexiconKit.


// Instantiate LexiconKit
NLexiconKit _LexiconKit = new NLexiconKit();
if (_LexiconKit != null)
{
    // Set credentials
    _LexiconKit.SeCredentials("Credentials");
    NSAPI5Synthesizer _Synthesizer = _LexiconKit.CreateSAPI5Synthesizer();
    if (_Synthesizer != null)
    {
        string phonemes = _Synthesizer.GeneratePhonemes("tomato", "Noun", "en-US", "sapi");
        _Synthesizer.SpeakPhonemes(phonemes, "en-US", "sapi");
        _Synthesizer.Dispose();
    }
    _LexiconKit.Dispose();
}


// Instantiate LexiconKit object
CLexiconKit* _LexiconKit = new CLexiconKit();
if (_LexiconKit != NULL)
{
	// Set credentials
	_LexiconKit->SetCredentials(L"Credentials");
	// Create synthesizer
	CSAPI5Synthesizer* _Synthesizer = _LexiconKit->CreateSAPI5Synthesizer();
	if (_Synthesizer != NULL)
	{
        wchar_t* phonemes = _Synthesizer->GeneratePhonemes(L"tomato", L"Noun", L"en-US", L"sapi");
        _Synthesizer->SpeakPhonemes(phonemes, L"en-US", L"sapi");
        delete _Synthesizer;
    }
    delete _LexiconKit;
}


// Instantiate LexiconKit  object
CLexiconKit* _LexiconKit = new CLexiconKit();
if (_LexiconKit != NULL)
{
    // Set credentials
    _LexiconKit->SetCredentials("Credentials");
    // Create synthesizer
    CSAPI5Synthesizer* _Synthesizer = _LexiconKit->CreateSAPI5Synthesizer();
    if (_Synthesizer != NULL)
    {
        String phonemes = _Synthesizer->GeneratePhonemes("tomato", "Noun", "en-US", "sapi");
        _Synthesizer->SpeakPhonemes(phonemes, "en-US", "sapi");
        delete _Synthesizer;
    }
    delete _LexiconKit;
}


var 
    _LexiconKit: TLexiconKit;
    _Synthesizer: TSAPI5Synthesizer;
    phonemes: string;
begin
    // Instantiate LexiconKit object
    _LexiconKit := TLexiconKit.Create();
    if (_LexiconKit <> nil) then
    begin
        // Set credentials
        _LexiconKit.SetCredentials('Credentials');
        // Create synthesizer
        _Synthesizer := _LexiconKit.CreateSAPI5Synthesizer();
        if (_Synthesizer <> nil) then
        begin
          phonemes := _Synthesizer.GeneratePhonemes('tomato', 'Noun', 'en-US', 'sapi');
          _Synthesizer.SpeakPhonemes(phonemes, 'en-US', 'sapi');
          _Synthesizer.Destroy();
        end;
        _LexiconKit.Destroy();
    end;
end;


JLexiconKit _LexiconKit = new JLexiconKit();
if (_LexiconKit != null)
{
	// Set credentials
	_LexiconKit.setCredentials("Credentials");
	JSAPI5Synthesizer _Synthesizer = _LexiconKit.createSAPI5Synthesizer();
	if (_Synthesizer != null)
	{
        String phonemes = _Synthesizer.generatePhonemes("tomato", "Noun", "en-US", "sapi");
        phonemes = _Synthesizer.speakPhonemes(phonemes, "en-US", "sapi");
        _Synthesizer.dispose();
	}
    _LexiconKit.dispose();
}


Dim _LexiconKit As NLexiconKit
Dim WithEvents _Synthesizer As NSAPI5Synthesizer
Dim phonemes As String
' Instantiate LexiconKit
_LexiconKit = New NLexiconKit()
If (_LexiconKit IsNot Nothing) Then
    ' Set credentials
    _LexiconKit.SetCredentials("Credentials")
    _Synthesizer = _LexiconKit.CreateSAPI5Synthesizer()
    If (_Synthesizer IsNot Nothing) Then
        phonemes = _Synthesizer.GeneratePhonemes("tomato", "Noun", "en-US", "sapi")
        _Synthesizer.SpeakPhonemes(phonemes, "en-US", "sapi")
        _Synthesizer.Dispose()
    End If
    _LexiconKit.Dispose()
End If

All synthesizers support speaking phonemes. Acapela TTS synthesizers and Microsoft WindowsMedia recognizers and synthesizers currently do not support generating phonemes.

Editing Pronunciations

To provide alternate pronunciations, edit lexicon word pronunciations (i.e., change the phonemes) and add them to the lexicon.

Speech engines (i.e., recognizers and synthesizers) support unique lexicon alphabets, formats, and approaches for runtime inclusion.

Speech API	Alphabets	File Format
Acapela TTS	ipa, acatts	.dic
class="px-2" Swift API	swift	.txt
Microsoft (SAPI5, Speech Platform, WindowsMedia)	ipa, sapi, ups	W3C .pls
Microsoft Azure Speech	ipa, sapi, ups, x-sampa	(W3C) .pls

Alphabets

A speech engine may support one or more alphabets. This varies by speech language. To determine which phoneme alphabets are support by an engine, enumerate the PhonemesAlphabets property. Use the applicable alphabet for generating, editing, and speaking alphabets.


foreach (string alphabet in _Synthesizer.PhonemeAlphabets)
{
    ...
}


for (int i = 0; i < _Synthesizer->GetPhonemeAlphabets()->GetCount(); i++)
{
    wchar_t* alphabet = _Synthesizer->GetPhonemeAlphabets()->at(i).c_str();
    ...
}


for (int i = 0; i < _Synthesizer->GetChantEngines()->Count; i++)
{
    String alphabet =  _Synthesizer->GetPhonemeAlphabets()[i];
    ...
}


var
alphabet: string;
begin
    for i := 0 to _Synthesizer.PhonemeAlphabets.Count-1 do
    begin
        // Access engine properties
        alphabet := _Synthesizer.PhonemeAlphabets[i]
        ...
    end;
end;


for (String alphabet : _Synthesizer.getPhonemeAlphabets())
{
    ...
}


For Each alphabet In _Synthesizer.PhonemeAlphabets
    ...
Next

In cases where a speech engine support multiple spoken languages (Azure), set the alphabetlanguage property to the language value before enumerating the PhonemeAlphabets property, generating, editing, and speaking phonemes.


// Set the synthesizer language
_Synthesizer.SetProperty("alphabetlanguage","de-DE");


// Set the synthesizer language
_Synthesizer->SetProperty(L"alphabetlanguage","50");


// Set the synthesizer language
_Synthesizer->SetProperty(L"alphabetlanguage","50");


// Set the synthesizer language
_Synthesizer.SetProperty('alphabetlanguage','50');


// Set the synthesizer language
_Synthesizer.setProperty(alphabetlanguage","50");


' Set the synthesizer language
_Synthesizer.SetProperty("alphabetlanguage","50")

Acapela TTS Properties

(Source: Acapela Group Acapela TTS Developer Guide)

Property	Value
enginepath	Runtime library path.
license	License file path.
preset	Name of the equalizer preset
lexicon	A list of user lexicons. Each file name must be separated with a semicolon.
pitch	The baseline pitch expressed in Hz. Man (110) Woman (180). Value ranges between 30 and 500.
speed	The reading speed. Values are in percent of the default speed rate 100. Value ranges between 30 and 300.
maxpitch	The maximum pitch allowed expressed in Hz. Value ranges between pitch and pitch * 2.5.
minpitch	The minimum pitch allowed expressed in Hz. Value ranges between pitch / 5 and pitch.
volume	A ratio percentage of the TTS output volume. Default value is 100. Value ranges 0 to 150.
leadingsilence	The pause duration at the beginning of speaking in milliseconds. Default value is 50. Value ranges 20 to 5000.
trailingsilence	The pause duration at the end of speaking in milliseconds. Default value is 500. Value ranges 20 to 5000.
deviceid	The audio device identifier.
readingmode	The way text is spoken: normal, word at a time, or letter at a time (spelling).
pausepunct	The pause duration for period, exclamation point, and question mark. Value ranges from 0 to 5.
pausesemicolon	The pause duration for semicolon. Value ranges from 0 to 5.
pausecomma	The pause duration for comma and colon. Value ranges from 0 to 5.
pausebracket	The pause duration for quote, braces, and brackets. Value ranges from 0 to 5.
pausespell	The pause duration between letters in spell reading mode. Value ranges from 0 to 5.
usefilter	Enable or disable equalizer use: 0 (no) or 1 (yes).
filtervalue1	A value corresponding to the attenuation for band 1 filter. Value ranges 0 to 200.
filtervalue2	A value corresponding to the attenuation for band 2 filter. Value ranges 0 to 200.
filtervalue3	A value corresponding to the attenuation for band 3 filter. Value ranges 0 to 200.
filtervalue4	A value corresponding to the attenuation for band 4 filter. Value ranges 0 to 200.
vocaltract	The voice shaping (tone) expressed in percentage of the default value 100. Value ranges 50 to 150.
audioboostpreemph	Controls the emphasis of medium and high frequencies. The default value is 0. Value ranges 0 to 90.

Cepstral Swift Properties

(Source: Cepstral Swift SDK documentation)

Property	Value
configfile	Configuration file to use instead of the default.
audiochannels	Number of audio channels [ 1 (mono) or 2 (stereo) ].
audiodeadair	Milliseconds of dead air (silence) to pad at the end of speech.
audiopan	Left-to-Right panning [ -1 = left, 0 = center, 1 = right ]. This must be used with audio/channels=2.
audiovolume	Volume multiplication factor as a percentage. Default value is 100.
lexicon	A lexicon.txt file in the voice directory. Review Ceptral Lexicon Editing with [LexiconKit](xref:lexiconkit "LexiconKit")
speechrate	Speaking rate (average WPM).
voicedir	Directory for the voice.
sfx	Special effects output chain file.

CereProc CereVoice Properties

(Sourcce: CereProc CereVoicec SDK User Guide)

Property	Value
enginepath	Runtime library path.
voicepath	Voice files path.
configfile	Configuration file path.
license	License file path.
rootcertificate	Root certificate file path.
clientcrt	Client CRT file path.
clientkey	Client key file paht.

Microsoft Azure Speech Properties

(Source: learn.microsoft.com)

Property	Value
alphabetlanguage	The voice language to use for synthesizers that support multiple languages from which to obtain supported phoneme alphabets.
devicename	The multimedia device ID that is used by the audio object.
language	The voice language to use for synthesizers that support multiple languages.
languages	One or more languages for which to enumerate available voices.
speechkey	The Azure Speech Services key.
speechregion	The The Azure Speech Services region.

Microsoft SAPI 5 Properties

(Source: Microsoft SAPI5 Help File)

Property	Value
deviceid	The multimedia device ID that is used by the audio object.
rate	The current text rendering rate adjustment. Value specifying the speaking rate of the voice. Supported values range from -10 to 10 - values outside this range may be truncated.
volume	The synthesizer output volume level of the voice in real time. Volume levels are specified in percentage values ranging from zero to 100. The default base volume for all voices is 100.

Perisisting Pronunciations

Once the lexicon containing lexicon word pronunciations is saved, it can be deployed it with applications and installed on a target system.

Acapela and Cepstral lexicons are loaded at run time by the SpeechKit class. See the speech synthesis Lexicons topic for details. W3C .pls lexicons are included as part of text-to-speech (TTS) markup. See VoiceMarkupKit for details.