Memories
Memories
What are Memories, and how do they work?
What are Memories, and how do they work?
Do Memories work across conversations?
Do Memories work across conversations?
What's the primary benefit of Memories?
What's the primary benefit of Memories?
How do I enable Memories in the UI?
How do I enable Memories in the UI?
How do I enable Memories via the API?
How do I enable Memories via the API?
memory_stores
field in the Create Conversation API call. This should be a stable, unique identifier for the user (e.g. user email, CRM ID, etc.). Example:Can I see or edit the Memories data?
Can I see or edit the Memories data?
Are Memories required for every conversation?
Are Memories required for every conversation?
Can multiple participants share the same Memories?
Can multiple participants share the same Memories?
What about customers who have already built their own Memories?
What about customers who have already built their own Memories?
In a situation where a user says, 'I said ABC,' but the AI Persona responded with 'DEF,' how can we investigate what was actually stored in Memories and understand why the AI Persona produced that response?
In a situation where a user says, 'I said ABC,' but the AI Persona responded with 'DEF,' how can we investigate what was actually stored in Memories and understand why the AI Persona produced that response?
For how long do Memories persist between interactions?
For how long do Memories persist between interactions?
Where can I find more information about Memories?
Where can I find more information about Memories?
Knowledge Base
Knowledge Base
What is Knowledge Base and how does it work?
What is Knowledge Base and how does it work?
What happens during a conversation when using RAG?
What happens during a conversation when using RAG?
- Analyzes the conversation context
- Retrieves relevant information from your document base
- Augments the AI’s responses with this contextual knowledge from your documents
How long does it take for knowledge to be retrieved?
How long does it take for knowledge to be retrieved?
What about customers who have already built their own Knowledge Base?
What about customers who have already built their own Knowledge Base?
Can you give an example of a Knowledge Base use case?
Can you give an example of a Knowledge Base use case?
What's the customer value of Knowledge Base?
What's the customer value of Knowledge Base?
- Faster onboarding (just upload the docs)
- More trustworthy answers, especially in regulated or high-stakes environments
- Higher task completion for users, thanks to grounded knowledge
What formats are supported for upload?
What formats are supported for upload?
- CSV
- TXT
- PPTX
- PNG
- JPG
- You can also enter any site URL and the Tavus API will scrape the site’s contents and reformat the content as a machine readable document.
Where can I find more information about Knowledge Base?
Where can I find more information about Knowledge Base?
Are there any access limitations when using a document ID?
Are there any access limitations when using a document ID?
How do I use Knowledge Base through the API?
How do I use Knowledge Base through the API?
How do I upload documents using API?
How do I upload documents using API?
What error codes might users encounter when uploading, accessing, or deleting documents, and what do they mean?
What error codes might users encounter when uploading, accessing, or deleting documents, and what do they mean?
file_size_too_large
– File exceeds the maximum allowed upload size.file_format_unsupported
– This file type isn’t supported for upload.invalid_file_url
– Provided file link is invalid or inaccessible.file_empty
– The uploaded file contains no readable content.website_processing_failed
– Website content could not be retrieved or processed.chunking_failed
– System couldn’t split file into processable parts.embedding_failed
– Failed to generate embeddings for your file content.vector_store_failed
– Couldn’t save data to the vector storage system.s3_storage_failed
– Error storing file in S3 cloud storage.contact_support
– An error occurred; please reach out for help.
How can I check the logs to see which document an AI persona referenced in a response to a customer?
How can I check the logs to see which document an AI persona referenced in a response to a customer?
What are the document_retrieval_strategy options?
What are the document_retrieval_strategy options?
- Speed: Optimizes for faster retrieval times for minimal latency.
- Balanced (default): Provides a balance between retrieval speed and quality.
- Quality: Prioritizes finding the most relevant information, which may take slightly longer but can provide more accurate responses.
How long does it take for documents to be uploaded and usable?
How long does it take for documents to be uploaded and usable?
Do we support all languages for the knowledge base?
Do we support all languages for the knowledge base?
Objectives and Guardrails
Objectives and Guardrails
Objectives - What are they? How do they work?
Objectives - What are they? How do they work?
How do I add Objectives to my Persona?
How do I add Objectives to my Persona?
What use cases are Objectives good for?
What use cases are Objectives good for?
Where can I find more information about Objectives?
Where can I find more information about Objectives?
Guardrails - What are they? How do they work?
Guardrails - What are they? How do they work?
How do I add Guardrails to my persona?
How do I add Guardrails to my persona?
Can I create different Guardrails for different Personas?
Can I create different Guardrails for different Personas?
Where can I find more information about Guardrails?
Where can I find more information about Guardrails?
Persona Builder
Persona Builder
What is the Persona Builder?
What is the Persona Builder?
How is this different from the old setup experience?
How is this different from the old setup experience?
General Tavus Q&A
General Tavus Q&A
What is Daily?
What is Daily?
Do I need a Daily account?
Do I need a Daily account?
- You do not need to sign up for a Daily account to use Tavus’s Conversational Video Interface.
- All you need is the Daily room URL (called
conversation_url
in our system) that is returned by the Tavus API. You can serve this link directly to your end users or embed it.
How do I embed the conversation using Daily's Prebuilt UI?
How do I embed the conversation using Daily's Prebuilt UI?
conversation_url
) ready, replace DAILY_ROOM_URL
in the code snippet below with your room URL.How do I embed the conversation using an iframe?
How do I embed the conversation using an iframe?
conversation_url
) ready, replace YOUR_TAVUS_MEETING_URL
in the iframe code snippet below with your room URL.How can I add custom LLM layers?
How can I add custom LLM layers?
layers
field when creating a persona using the Create Persona API. Example configuration:How do I modify TTS voices?
How do I modify TTS voices?
- Cartesia
- ElevenLabs
How do I add call-back URLs?
How do I add call-back URLs?
callback_url
property when creating the conversationHow do I get transcripts for my conversation?
How do I get transcripts for my conversation?
POST
requests from Tavus. This endpoint will receive the callback events for the transcripts after the conversation ended. Then, add callback_url
property when creating the conversation.event_type = application.transcription_ready
when the transcript is ready.How do I get visual summary for my conversation?
How do I get visual summary for my conversation?
POST
requests from Tavus. This endpoint will receive the callback events for the visual summary after the conversation ended. Then, add callback_url
property when creating the conversation.event_type = application.perception_analysis
when the summary is ready.What LLM (Large Language Model) does Tavus use to power the conversational replicas?
What LLM (Large Language Model) does Tavus use to power the conversational replicas?
-
Tavus-Provided LLMs: You can choose between three different models:
tavus-llama-4
: The default choice if no LLM layer is provided. This is the smartest and fastest model, offering the best user-to-user (U2U) experience. It’s on-premise, making it incredibly performant.tavus-gpt-4o
: Another viable option for complex interactions.tavus-gpt-4o-mini
: Faster thantavus-gpt-4o
at the slight cost of performance.
- No LLM Layer: If you don’t include an LLM layer, Tavus will automatically default to a Tavus-provided model.
What is the maximum context window supported by the default LLM?
What is the maximum context window supported by the default LLM?
- The default LLM,
tavus-llama-4
, has a limit of 32,000 tokens. - Contexts over 25,000 tokens will experience noticeable performance degradation (slower response times).
What are some recording tips for producing high quality conversational replica training footage?
What are some recording tips for producing high quality conversational replica training footage?
- Minimal Head Movement: Aim to keep your head and body as still as possible during the recording. This helps in maintaining consistency and improves the overall quality of the training data.
- Pause and Be Still: It’s recommended to stop, stay still, and remain silent for at least 5 seconds at regular intervals throughout the script. These pauses are crucial for helping the replica appear natural during moments of silence in a conversation.
- Use a Laptop Camera: Recording on a laptop camera, as if you were on a Zoom call, often yields the most natural results. This setup mimics a familiar conversational setting, enhancing the naturalness of the footage.
How do I add perception tool calls?
How do I add perception tool calls?
layers.perception
object when creating a persona:Do I need to invite the replica to the meeting room?
Do I need to invite the replica to the meeting room?
For CVI, what's customizable vs. out of the box?
For CVI, what's customizable vs. out of the box?
How does Tavus deliver real-time responsiveness?
How does Tavus deliver real-time responsiveness?
What's behind the scenes of CVI?
What's behind the scenes of CVI?
- LLMs for natural language understanding
- Real-time rendering for facial video
- APIs for Persona creation and conversational control
How many languages does Tavus support?
How many languages does Tavus support?
Can Tavus support different accents or dialects?
Can Tavus support different accents or dialects?
What can Tavus do when it comes to orchestration (calendars, email tools, HubSpot, DocuSign, etc.)?
What can Tavus do when it comes to orchestration (calendars, email tools, HubSpot, DocuSign, etc.)?
What makes a good prompt? How much does Tavus help with that?
What makes a good prompt? How much does Tavus help with that?
How do I add a custom LLM to CVI?
How do I add a custom LLM to CVI?
How customizable is the user interface? What does Tavus provide?
How customizable is the user interface? What does Tavus provide?
How do I change the AI Persona's voice?
How do I change the AI Persona's voice?
How can I reduce background noise during calls?
How can I reduce background noise during calls?
Can I track events in the video call?
Can I track events in the video call?
How do you change or customize backgrounds or AI Persona?
How do you change or customize backgrounds or AI Persona?
What compliance and security standards does Tavus meet?
What compliance and security standards does Tavus meet?
- SOC 2 compliant
- GDPR compliant
- HIPAA compliant
- BAA compliant
How does Tavus protect customer data and support data privacy?
How does Tavus protect customer data and support data privacy?
- We never train our models on your data.
- Any data processed through Tavus is not retained by default.
- Conversations and Personas are managed using anony/mized IDs.
- You own and store all transcripts, conversation recordings, and outputs.
- Memory and knowledge base are stored internally and only utilize specific datasets related to the conversation.