Advanced Prompt Configuration

The Switchboard framework provides powerful prompt capabilities including text-to-speech, SSML, audio files, and multi-language support.

Overview

Prompts can be configured in two ways:

Simple Text Prompts - Quick, inline text strings
Advanced Configuration - Full control over voice, language, SSML, audio files, and more

Simple Text Prompts

For basic prompts, use the simple string overload:

csharp

.PlayPrompt("Welcome to our service")
.PlayPrompt("Thank you for calling")

Advanced Prompt Configuration

For advanced scenarios, use the configuration lambda:

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Text;
    prompt.Text = "Welcome to our service";
    prompt.Voice = "Matthew";
    prompt.LanguageCode = "en-US";
    prompt.UseNeuralVoice = true;
})

Prompt Types

The framework supports four prompt types:

1. Text (Default)

Plain text converted to speech using Amazon Polly:

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Text; // Default
    prompt.Text = "Welcome to customer service";
    prompt.Voice = "Joanna"; // US English female voice
})

2. SSML (Speech Synthesis Markup Language)

Advanced speech control with pauses, emphasis, pronunciation, and more:

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = @"<speak>
        Welcome to our service.
        <break time='1s'/>
        How may I <emphasis level='strong'>assist</emphasis> you today?
    </speak>";
    prompt.Voice = "Matthew";
    prompt.UseNeuralVoice = true;
})

3. Audio Files (S3)

Pre-recorded audio files stored in Amazon S3:

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Audio;
    prompt.S3BucketName = "my-prompts-bucket";
    prompt.S3Key = "welcome-message.wav";
})

4. Library Prompts

Amazon Connect prompt library prompts:

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.LibraryPrompt;
    prompt.AudioPromptArn = "arn:aws:connect:us-east-1:123456789012:instance/abc123/prompt/xyz789";
})

SSML Examples

Pauses and Breaks

csharp

prompt.SSML = @"<speak>
    Welcome to customer service.
    <break time='500ms'/>
    Please listen carefully to the following options.
</speak>";

Emphasis

csharp

prompt.SSML = @"<speak>
    This is <emphasis level='moderate'>important</emphasis>.
    Please <emphasis level='strong'>do not</emphasis> hang up.
</speak>";

Speaking Rate

csharp

prompt.SSML = @"<speak>
    <prosody rate='slow'>
        I will speak slowly for clarity.
    </prosody>
    <break time='1s'/>
    <prosody rate='fast'>
        Now I'll speak quickly.
    </prosody>
</speak>";

Volume Control

csharp

prompt.SSML = @"<speak>
    <prosody volume='soft'>This is soft.</prosody>
    <prosody volume='loud'>This is loud!</prosody>
</speak>";

Pronunciation

csharp

prompt.SSML = @"<speak>
    The chemical symbol is
    <say-as interpret-as='spell-out'>H2O</say-as>.
    Your confirmation number is
    <say-as interpret-as='digits'>1234567</say-as>.
</speak>";

Numbers and Dates

csharp

prompt.SSML = @"<speak>
    Your appointment is on
    <say-as interpret-as='date' format='mdy'>12/25/2025</say-as>
    at
    <say-as interpret-as='time'>2:30pm</say-as>.
</speak>";

Voice Selection

Popular US English Voices

csharp

// Female voices
prompt.Voice = "Joanna";  // US English, warm and friendly
prompt.Voice = "Kendra";  // US English, professional
prompt.Voice = "Kimberly"; // US English, conversational
prompt.Voice = "Salli";   // US English, clear
prompt.Voice = "Ivy";     // US English, young adult

// Male voices
prompt.Voice = "Matthew"; // US English, professional
prompt.Voice = "Joey";    // US English, young adult
prompt.Voice = "Justin";  // US English, young adult

International Voices

csharp

// British English
prompt.Voice = "Amy";    // British English female
prompt.Voice = "Emma";   // British English female
prompt.Voice = "Brian";  // British English male

// Canadian French
prompt.Voice = "Chantal"; // Canadian French female

// Mexican Spanish
prompt.Voice = "Mia";    // Mexican Spanish female

// Brazilian Portuguese
prompt.Voice = "Camila"; // Brazilian Portuguese female, neural

Complete Voice List

For a complete list of available voices, see: Amazon Polly Voice List

Neural Voices

Neural voices provide more natural, human-like speech:

csharp

.PlayPrompt(prompt =>
{
    prompt.Text = "Welcome to our premium service";
    prompt.Voice = "Joanna";
    prompt.UseNeuralVoice = true; // Enable neural engine
})

Neural Voice Benefits:

More natural intonation and rhythm
Better prosody (stress and timing)
Smoother speech quality
Higher cost (check AWS pricing)

Availability:

Not all voices support neural mode
Check voice compatibility in AWS documentation
Falls back to standard if neural unavailable

Multi-Language Support

Language Code Configuration

csharp

.PlayPrompt(prompt =>
{
    prompt.Text = "Bienvenue à notre service";
    prompt.Voice = "Celine";
    prompt.LanguageCode = "fr-CA"; // Canadian French
    prompt.UseNeuralVoice = true;
})

Language-Specific Examples

Spanish (US):

csharp

.PlayPrompt(prompt =>
{
    prompt.Text = "Bienvenido a nuestro servicio de atención al cliente";
    prompt.Voice = "Lupe";  // US Spanish neural
    prompt.LanguageCode = "es-US";
    prompt.UseNeuralVoice = true;
})

French (Canadian):

csharp

.PlayPrompt(prompt =>
{
    prompt.Text = "Bienvenue à notre service client";
    prompt.Voice = "Chantal";
    prompt.LanguageCode = "fr-CA";
})

Portuguese (Brazilian):

csharp

.PlayPrompt(prompt =>
{
    prompt.Text = "Bem-vindo ao nosso serviço de atendimento ao cliente";
    prompt.Voice = "Camila";
    prompt.LanguageCode = "pt-BR";
    prompt.UseNeuralVoice = true;
})

Audio File Prompts

S3 Audio Configuration

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Audio;
    prompt.S3BucketName = "my-company-prompts";
    prompt.S3Key = "greetings/welcome-en.wav";
})

Audio File Requirements

Supported Formats:

WAV (recommended): 8 kHz, 16-bit, mono
MP3: Up to 48 kHz

Best Practices:

Use WAV for best quality
8 kHz sample rate for telephony
16-bit depth
Mono (single channel)
Keep files under 50 MB
Use consistent audio levels

S3 Bucket Setup

bash

# Create S3 bucket for prompts
aws s3 mb s3://my-company-prompts

# Upload audio file
aws s3 cp welcome.wav s3://my-company-prompts/greetings/

# Set bucket policy for Amazon Connect access
# (See AWS Connect documentation for IAM policy)

Complex SSML Examples

Professional IVR Greeting

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = @"<speak>
        Thank you for calling
        <emphasis level='moderate'>Acme Corporation</emphasis>.
        <break time='500ms'/>
        Your call may be monitored or recorded for quality assurance purposes.
        <break time='1s'/>
        Please listen carefully as our menu options have changed.
    </speak>";
    prompt.Voice = "Joanna";
    prompt.UseNeuralVoice = true;
})

After Hours Message

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = @"<speak>
        Thank you for calling.
        <break time='500ms'/>
        Our office is currently <emphasis level='strong'>closed</emphasis>.
        <break time='1s'/>
        Our business hours are
        <prosody rate='slow'>
            Monday through Friday, 9 AM to 5 PM Eastern Time.
        </prosody>
        <break time='1s'/>
        Please call back during business hours,
        or visit our website at
        <say-as interpret-as='spell-out'>www</say-as> dot acme dot com.
    </speak>";
    prompt.Voice = "Matthew";
})

Account Information Readback

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = $@"<speak>
        Your account number is
        <say-as interpret-as='digits'>{accountNumber}</say-as>.
        <break time='1s'/>
        Your current balance is
        <say-as interpret-as='cardinal'>{balance}</say-as>
        dollars and
        <say-as interpret-as='cardinal'>{cents}</say-as>
        cents.
    </speak>";
    prompt.Voice = "Kendra";
})

Dynamic Prompts

Context-Based Greetings

csharp

// Time-based greeting
var greeting = DateTime.Now.Hour < 12 ? "Good morning" :
               DateTime.Now.Hour < 17 ? "Good afternoon" :
               "Good evening";

.PlayPrompt(prompt =>
{
    prompt.Text = $"{greeting}, welcome to customer service";
    prompt.Voice = "Joanna";
})

Personalized Messages

csharp

// Using contact attributes
.SetContactAttributes(attrs =>
{
    attrs["CustomerName"] = "$.External.CustomerName";
    attrs["AccountType"] = "$.External.AccountType";
})
.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = @"<speak>
        Welcome back
        <emphasis level='moderate'>
            <say-as interpret-as='name'>$.Attributes.CustomerName</say-as>
        </emphasis>.
        <break time='500ms'/>
        I see you have a $.Attributes.AccountType account.
    </speak>";
})

Validation

The framework automatically validates prompt configuration:

csharp

// ✅ Valid: Text provided
.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Text;
    prompt.Text = "Welcome";
})

// ❌ Invalid: Missing text
.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Text;
    // Throws: "Text is required when PromptType is Text"
})

// ❌ Invalid: Invalid SSML
.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = "Missing speak tags";
    // Throws: "SSML must be wrapped in <speak> tags"
})

// ❌ Invalid: Missing S3 info
.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Audio;
    prompt.S3BucketName = "my-bucket";
    // Throws: "S3BucketName and S3Key are required when PromptType is Audio"
})

Best Practices

1. Use Appropriate Prompt Types

csharp

// Simple messages: Use Text
.PlayPrompt("Thank you for calling")

// Complex messages: Use SSML
.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = "<speak>Your payment of <say-as interpret-as='currency'>$100.50</say-as> has been received</speak>";
})

// Branding/consistency: Use Audio
.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Audio;
    prompt.S3BucketName = "company-branding";
    prompt.S3Key = "welcome-message.wav";
})

2. Voice Consistency

csharp

// Pick one voice and stick with it throughout the flow
const string VOICE = "Joanna";

.PlayPrompt(prompt => { prompt.Voice = VOICE; prompt.Text = "Welcome"; })
.PlayPrompt(prompt => { prompt.Voice = VOICE; prompt.Text = "Thank you"; })

3. Neural Voices for Customer Experience

csharp

// Use neural voices for customer-facing prompts
.PlayPrompt(prompt =>
{
    prompt.Text = "Welcome valued customer";
    prompt.Voice = "Joanna";
    prompt.UseNeuralVoice = true; // More natural sound
})

// Use standard voices for internal/testing
.PlayPrompt(prompt =>
{
    prompt.Text = "Test environment - agent callback";
    prompt.Voice = "Matthew";
    prompt.UseNeuralVoice = false; // Lower cost
})

4. Accessibility

csharp

// Use clear, slow speech for important information
.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = @"<speak>
        <prosody rate='slow'>
            Your confirmation code is
            <say-as interpret-as='spell-out'>A B C 1 2 3</say-as>.
        </prosody>
        <break time='2s'/>
        I'll repeat that.
        <prosody rate='slow'>
            <say-as interpret-as='spell-out'>A B C 1 2 3</say-as>.
        </prosody>
    </speak>";
})

5. Testing Different Voices

csharp

#if DEBUG
    const string VOICE = "Matthew"; // Fast to test
#else
    const string VOICE = "Joanna";  // Production voice
    const bool NEURAL = true;
#endif

Common Patterns

Multi-Step Confirmation

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = @"<speak>
        I heard you say
        <emphasis level='strong'>cancel my subscription</emphasis>.
        <break time='1s'/>
        Press 1 to confirm, or press 2 to return to the main menu.
    </speak>";
    prompt.Voice = "Kendra";
})

Hold Music Simulation

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.Audio;
    prompt.S3BucketName = "company-audio";
    prompt.S3Key = "hold-music/jazz-loop.mp3";
})

Bilingual Greeting

csharp

.PlayPrompt(prompt =>
{
    prompt.PromptType = PromptType.SSML;
    prompt.SSML = @"<speak>
        Welcome to customer service.
        <break time='500ms'/>
        Para español, oprima dos.
        <break time='500ms'/>
        For English, press one.
    </speak>";
    prompt.Voice = "Joanna";
})

Input Handling - Customer input configuration
Flow Basics - Fundamental flow concepts
SSML Reference - Complete SSML documentation
Amazon Polly Voices - Available voices

Advanced Prompt Configuration ​

Overview ​

Simple Text Prompts ​

Advanced Prompt Configuration ​

Prompt Types ​

1. Text (Default) ​

2. SSML (Speech Synthesis Markup Language) ​

3. Audio Files (S3) ​

4. Library Prompts ​

SSML Examples ​

Pauses and Breaks ​

Emphasis ​

Speaking Rate ​

Volume Control ​

Pronunciation ​

Numbers and Dates ​

Voice Selection ​

Popular US English Voices ​

International Voices ​

Complete Voice List ​

Neural Voices ​

Multi-Language Support ​

Language Code Configuration ​

Language-Specific Examples ​

Audio File Prompts ​

S3 Audio Configuration ​

Audio File Requirements ​

S3 Bucket Setup ​

Complex SSML Examples ​

Professional IVR Greeting ​

After Hours Message ​

Account Information Readback ​

Dynamic Prompts ​

Context-Based Greetings ​

Personalized Messages ​

Validation ​

Best Practices ​

1. Use Appropriate Prompt Types ​

2. Voice Consistency ​

3. Neural Voices for Customer Experience ​

4. Accessibility ​

5. Testing Different Voices ​

Common Patterns ​

Multi-Step Confirmation ​

Hold Music Simulation ​

Bilingual Greeting ​

Related Documentation ​

Advanced Prompt Configuration

Overview

Simple Text Prompts

Advanced Prompt Configuration

Prompt Types

1. Text (Default)

2. SSML (Speech Synthesis Markup Language)

3. Audio Files (S3)

4. Library Prompts

SSML Examples

Pauses and Breaks

Emphasis

Speaking Rate

Volume Control

Pronunciation

Numbers and Dates

Voice Selection

Popular US English Voices

International Voices

Complete Voice List

Neural Voices

Multi-Language Support

Language Code Configuration

Language-Specific Examples

Audio File Prompts

S3 Audio Configuration

Audio File Requirements

S3 Bucket Setup

Complex SSML Examples

Professional IVR Greeting

After Hours Message

Account Information Readback

Dynamic Prompts

Context-Based Greetings

Personalized Messages

Validation

Best Practices

1. Use Appropriate Prompt Types

2. Voice Consistency

3. Neural Voices for Customer Experience

4. Accessibility

5. Testing Different Voices

Common Patterns

Multi-Step Confirmation

Hold Music Simulation

Bilingual Greeting

Related Documentation