Use the Speak AI MCP server tool ask_ai_chat to ask an AI-powered question about
your media using Speak AI’s AI Chat.
What does it do?
Ask an AI-powered question about your media using Speak AI’s AI Chat. Supports
querying a single file, multiple files, entire folders, or your whole workspace.
Pass mediaIds for specific files, folderIds for entire folders, or omit both to
search across all media. Use assistantType to get specialized responses (e.g.,
‘researcher’ for academic analysis, ‘sales’ for deal insights). To continue a
conversation, pass the promptId from a previous response. Returns a promptId: save
it to continue the conversation with follow-up questions. Set analysisMediaId +
analysisInput to have the model listen to the audio or watch the video instead of
reading the transcript alone. That is a premium feature and costs credits per hour
of media: call get_analysis_quote first to check eligibility and price.
Parameters
ask_ai_chat takes 17 parameters, 1 of them required.
| Name | Type | Required | Description |
|---|---|---|---|
prompt |
string | Yes | The question or prompt to ask about the media |
mediaIds |
array of string | No | Array of media IDs to query. Omit along with folderIds to search across all media in your workspace. |
folderIds |
array of string | No | Array of folder IDs to scope the query to. Omit along with mediaIds to search across all media. |
folderId |
string | No | Single folder ID to scope the query to. Use folderIds for multiple folders. |
assistantType |
enum | No | Assistant persona: ‘general’ (default), ‘researcher’ (academic), ‘marketer’ (content), ‘sales’ (deals), ‘recruiter’ (hiring). Use ‘custom’ with assistantTemplateId. |
assistantTemplateId |
string | No | Required when assistantType is ‘custom’. ID of a custom assistant template from list_prompts. |
promptId |
string | No | ID of an existing conversation to continue. Pass this to maintain chat context across multiple questions. |
speakers |
array of string | No | Filter to specific speaker IDs from the transcript |
tags |
array of string | No | Filter media by tags |
startDate |
string | No | Start date for date range filter (ISO 8601, e.g., ‘2025-01-01’) |
endDate |
string | No | End date for date range filter (ISO 8601, e.g., ‘2025-03-31’) |
isIndividualPrompt |
boolean | No | When true, processes each media file separately instead of combining context. Useful for comparing responses across files. |
fieldId |
string | No | Scope the prompt to a single custom field |
fieldIds |
array of string | No | Scope the prompt to multiple custom fields (max 10) |
filters |
object | No | Advanced filter object to scope which media the prompt runs over |
analysisMediaId |
string | No | Media to analyse as audio/video rather than transcript. Must also appear in mediaIds, and must be sent together with analysisInput. Premium feature. |
analysisInput |
enum: audio, video | No | ‘audio’ lets the model hear tone, pacing and delivery; ‘video’ also lets it see what is on screen. Omit for transcript-only, which is the default and costs nothing extra. ‘transcript’ is not a valid value here: omitting the field IS transcript-only. |
Behaviour
The Speak AI MCP server publishes these behaviour hints for ask_ai_chat.
- Can change data in your workspace.
- Does not delete or overwrite existing data.
- Repeating an identical call can have an extra effect.
- Stays inside Speak AI. Does not reach other systems.
Related
- Ask AI Chat lists the rest of the Ask AI Chat category.
- Tool reference lists every Speak AI MCP server tool.
- MCP server covers setup, supported clients, and what agents can do once connected.