Struct SherpaOnnxKeywordResult

#[repr(C)]
pub struct SherpaOnnxKeywordResult {
    pub keyword: *const c_char,
    pub tokens: *const c_char,
    pub tokens_arr: *const *const c_char,
    pub count: i32,
    pub timestamps: *mut f32,
    pub start_time: f32,
    pub json: *const c_char,
}

Fields

keyword: *const c_char

The triggered keyword. For English, it consists of space separated words. For Chinese, it consists of Chinese words without spaces. Example 1: “hello world” Example 2: “你好世界”

tokens: *const c_char

Decoded results at the token level. For instance, for BPE-based models it consists of a list of BPE tokens.

tokens_arr: *const *const c_char

count: i32

timestamps: *mut f32

timestamps.size() == tokens.size() timestamps[i] records the time in seconds when tokens[i] is decoded.

start_time: f32

Starting time of this segment. When an endpoint is detected, it will change

json: *const c_char

Return a json string.

The returned string contains: { “keyword”: “The triggered keyword”, “tokens”: [x, x, x], “timestamps”: [x, x, x], “start_time”: x, }

Trait Implementations

impl Clone for SherpaOnnxKeywordResult

fn clone(&self) -> SherpaOnnxKeywordResult

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for SherpaOnnxKeywordResult

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Copy for SherpaOnnxKeywordResult