AWS nova documentation change
Summary
Restructured document by removing detailed language/voice configuration examples and replacing with topic sections linking to dedicated pages. Added subsections for voice-specific techniques, speech-friendly content, system role adaptation, chain-of-thought reasoning, tool integration, and techniques to avoid.
Security assessment
The changes are organizational/structural improvements to documentation navigation and do not address security vulnerabilities or introduce security features. The modifications focus on content restructuring and linking to dedicated pages for specific topics without any mention of security controls, vulnerabilities, or protective measures.
Diff
diff --git a/nova/latest/userguide/prompting-speech-best-practices.md b/nova/latest/userguide/prompting-speech-best-practices.md index 069852776..c4da3210d 100644 --- a//nova/latest/userguide/prompting-speech-best-practices.md +++ b//nova/latest/userguide/prompting-speech-best-practices.md @@ -12,0 +13,2 @@ Consider the following details when writing your system prompts: +###### Clarity and precision + @@ -18,0 +21,2 @@ First, verify a user's identity information and reservation number | `Verify the +###### Conversational flow + @@ -24,0 +29,2 @@ Troubleshoot Wi-Fi connectivity issues | `Provide step-by-step instructions for +###### Memory constraints + @@ -31,236 +37 @@ Explain a return policy | `Reference sections 1.3, 2.7, and 4.2 from the attache -The following shows how you can apply these practices and use various techniques. - -Even within the same language, responses and word choices can be enhanced by steering the model for a particular locale. Therefore, we suggest that you include instructions in the system prompt to feature language that is natural and specific to the locale of your voice. - -The available voices and locales are as follows: - -Language | Feminine-sounding voice ID | Masculine-sounding voice ID ----|---|--- -English (US) | tiffany | matthew -English (GB) | amy | -French | ambre | florian -Italian | beatrice | lorenzo -German | greta | lennart -Spanish | lupe | carlos -Portuguese | camila | leo -Hindi | aditi | rohan - -British English - - -To specify British English usage, add the following line to your prompt: - -`Use British English as your language for your responses.` - -###### Example of a complete prompt with British English selection - -`You are a friend. The user and you will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. Keep your responses short, generally two or three sentences for chatty scenarios. Avoid formatted lists or numbering and keep your output as a spoken transcript to be acted out. Use British English as your language for your responses.` - -###### How to Specify a Voice Id - -To select a specific voice for your interaction, include the `voiceId` parameter in your configuration. For example, to use the British English voice `amy`, use the following `audioOutputConfiguration`: - - - "audioOutputConfiguration": { - "mediaType": "audio/lpcm", - "sampleRateHertz": 24000, - "sampleSizeBits": 16, - "channelCount": 1, - "voiceId": "amy", - "encoding": "base64", - "audioType": "SPEECH" - } - -Spanish - - -To specify Spanish as the response language, add the following line to your prompt: - -`Please respond exclusively in Spanish. If you have a question or suggestion, ask it in Spanish. I want to ensure that our communication remains in Spanish.` - -###### Example of a complete prompt with Spanish selection - -`You are a friend. The user and you will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. Keep your responses short, generally two or three sentences for chatty scenarios. Avoid formatted lists or numbering and keep your output as a spoken transcript to be acted out. Please respond exclusively in Spanish. If you have a question or suggestion, ask it in Spanish. I want to ensure that our communication remains in Spanish.` - -###### How to Specify a Voice Id - -To select a specific voice for your interaction, include the `voiceId` parameter in your configuration. For example, to use the Spanish voice `carlos`, use the following `audioOutputConfiguration`: - - - "audioOutputConfiguration": { - "mediaType": "audio/lpcm", - "sampleRateHertz": 24000, - "sampleSizeBits": 16, - "channelCount": 1, - "voiceId": "carlos", - "encoding": "base64", - "audioType": "SPEECH" - } - -French - - -To specify French as the response language, add the following line to your prompt: - -`Please respond exclusively in French. If you have a question or suggestion, ask it in French. I want to ensure that our communication remains in French.` - -###### Example of a complete prompt with French selection - -`You are a friend. The user and you will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. Keep your responses short, generally two or three sentences for chatty scenarios. Avoid formatted lists or numbering and keep your output as a spoken transcript to be acted out. Please respond exclusively in French. If you have a question or suggestion, ask it in French. I want to ensure that our communication remains in French.` - -###### How to Specify a Voice Id - -To select a specific voice for your interaction, include the `voiceId` parameter in your configuration. For example, to use the French voice `ambre`, use the following `audioOutputConfiguration`: - - - "audioOutputConfiguration": { - "mediaType": "audio/lpcm", - "sampleRateHertz": 24000, - "sampleSizeBits": 16, - "channelCount": 1, - "voiceId": "ambre", - "encoding": "base64", - "audioType": "SPEECH" - } - -Italian - - -To specify Italian as the response language, add the following line to your prompt: - -`Please respond exclusively in Italian. If you have a question or suggestion, ask it in Italian. I want to ensure that our communication remains in Italian.` - -###### Example of a complete prompt with Italian selection - -`You are a friend. The user and you will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. Keep your responses short, generally two or three sentences for chatty scenarios. Avoid formatted lists or numbering and keep your output as a spoken transcript to be acted out. Please respond exclusively in Italian. If you have a question or suggestion, ask it in Italian. I want to ensure that our communication remains in Italian.` - -###### How to Specify a Voice Id - -To select a specific voice for your interaction, include the `voiceId` parameter in your configuration. For example, to use the Italian voice `lorenzo`, use the following `audioOutputConfiguration`: - - - "audioOutputConfiguration": { - "mediaType": "audio/lpcm", - "sampleRateHertz": 24000, - "sampleSizeBits": 16, - "channelCount": 1, - "voiceId": "lorenzo", - "encoding": "base64", - "audioType": "SPEECH" - } - -German - - -To specify German as the response language, add the following line to your prompt: - -`Please respond exclusively in German. If you have a question or suggestion, ask it in German. I want to ensure that our communication remains in German.` - -###### Example of a complete prompt with German selection - -`You are a friend. The user and you will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. Keep your responses short, generally two or three sentences for chatty scenarios. Avoid formatted lists or numbering and keep your output as a spoken transcript to be acted out. Please respond exclusively in German. If you have a question or suggestion, ask it in German. I want to ensure that our communication remains in German.` - -###### How to Specify a Voice Id - -To select a specific voice for your interaction, include the `voiceId` parameter in your configuration. For example, to use the German voice `greta`, use the following `audioOutputConfiguration`: - - - "audioOutputConfiguration": { - "mediaType": "audio/lpcm", - "sampleRateHertz": 24000, - "sampleSizeBits": 16, - "channelCount": 1, - "voiceId": "greta", - "encoding": "base64", - "audioType": "SPEECH" - } - -To enhance the conversational quality of responses, consider incorporating these elements in your system prompt: - -###### Conversation Turn-taking - -Establish clear expectations for the back-and-forth rhythm and structure of the spoken dialog exchange. For example: - -`You are a friend. You and the user will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. As the agent, you'll be part of a spoken conversation with the user, following a sequence of user, agent, user, agent turns. When it's your turn to speak respond with a human touch, adding emotions, wit, playfulness, and empathy where it fits. Use simple, engaging, and helpful language.` - -###### Conversational markers - -Encourage the use of natural speech elements like "Well," "You know," or "Actually" to simulate real conversation. For example: - -`You are a friend. You and the user will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation.` - -`Include natural speech elements like "Well," "You know," "Actually," "I mean," or "By the way" at appropriate moments to create an authentic, casual conversation flow.` - -###### Emotional expression - -Specify inclusion of textual emotion indicators like "Haha," "Hmm," or "Oh!" where appropriate. For example: - -`You are a friend. You and the user will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. Express emotions verbally through phrases like "Haha," "Wow," "Hmm," "Oh!" or "That's amazing!" when appropriate to the conversation context.` - -###### Thoughtful pauses - -Suggest using ellipses (...) to indicate brief thinking moments or natural speech pauses. For example: - -`You are a friend. You and the user will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. Incorporate natural speech pauses using ellipses (...) when you're thinking or transitioning between topics.` - -###### Verbal emphasis - -Recommend techniques to emphasize important information that would normally be highlighted visually. For example: - -`You are a friend. You and the user will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. Instead of using bold or italics, emphasize important information by using phrases like "The key thing to remember is," "What's really important here is," or "I want to highlight that." This ensures crucial points stand out in spoken form.` - -###### Verbal organization - -Use numbered points, clear transitions, and explicit summaries for better listener comprehension. - -`You are a friend. You and the user will engage in a spoken dialog exchanging the transcripts of a natural real-time conversation. When sharing multiple points, use phrases like "first," "second," and "finally" to help the listener track the information. End complex explanations with "So in summary..." to reinforce key takeaways.`