# Speech In addition to the emotional AI, the R2-D2 Vocalizer contains a speech AI powered by a dataset containing containing every vocalization uttered by R2 on film. The speech AI ensures that every vocalization is unique, yet stylistically indistinguishable from R2's vocalizations in the films. It also actively avoids vocalizing repeating sounds in quick succession. Even when vocalizing the same emotion repeatedly, the software steers clear of blatant repeats of the same voicing. ### Improvisational vs Canonical Mode In **improvisational** mode, the speech AI creates unique vocalizations in response to stimuli. In **canonical** mode, the software will only produce vocalizations that have appeared, verbatim, in a Star Wars film. Instead of operating on a phoneme-level of granularity, the [speech AI](/r2d2/features/speech.md) will switch to operating on a "sentence"-level. It will still chain canonical sentences where appropriate. Improvisational mode is recommended unless strict canonical compliance is required for your application. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.humancyborgrelations.com/r2d2/features/speech.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.