Struct Transcript 

pub struct Transcript {
    pub text: String,
    pub asr: String,
    pub stages: Stages,
    pub truncated: bool,
}

Clean text, plus what it was before normalization.

Fields

text: String

The text a caller should use.

Not necessarily one paragraph. Long input produces paragraph breaks even under the defaults, and crate::normalize::Structure::Lists and crate::normalize::Context::Email make multi-line output likely - bullets and greeting/sign-off blocks respectively. A fixed-height single-line surface will notice.

asr: String

Recognition output before normalization. Equal to text when no normalizer ran. Deliberately not named raw, which the recognition library already uses for something else.

stages: Stages

Where the time went.

truncated: bool

This is what fitted rather than what was said: the recording hit Config::max_capture, the decode ran out of its output budget, a window stopped on a repetition loop, or a window failed after earlier windows produced words. Carried here rather than only logged: a host cannot tell from a log line it may not be routing that part of the take is missing.

Which cause it was depends on the path. Audio that arrived through Engine::transcribe or Engine::transcribe_with was never capped, so a cap is not among its causes; on the capture path a host can compare Stages::audio against its configured cap to tell the cap from the rest.

Trait Implementations

impl Clone for Transcript

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Transcript

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

Formats the value using the given formatter.

impl PartialEq for Transcript

fn eq(&self, other: &Self) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for Transcript