Enum TimestampMode 

pub enum TimestampMode {
    Tokens,
    Words,
    Sentences,
}

Timestamp output mode for transcription results

Determines how token-level timestamps are grouped and presented:

• Tokens: Raw token-level output from the model (most detailed)
• Words: Tokens grouped into individual words
• Sentences: Tokens grouped by sentence boundaries (., ?, !)

Model-Specific Recommendations

• Parakeet CTC (English): Use Words mode. The CTC model only outputs lowercase alphabet without punctuation, so sentence segmentation is not possible.
• Parakeet TDT (Multilingual): Use Sentences mode. The TDT model predicts punctuation, enabling natural sentence boundaries.

Variants

Tokens

Raw token-level timestamps from the model

Words

Word-level timestamps (groups subword tokens)

Sentences

Sentence-level timestamps (groups by punctuation)

Note: Only works with models that predict punctuation (e.g., Parakeet TDT). CTC models don’t predict punctuation, so use Words mode instead.

Trait Implementations

impl Clone for TimestampMode

fn clone(&self) -> TimestampMode

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TimestampMode

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

Formats the value using the given formatter.

impl Default for TimestampMode

fn default() -> TimestampMode

Returns the “default value” for a type.

impl PartialEq for TimestampMode

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

Tests for self and other values to be equal, and is used by ==.

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

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Copy for TimestampMode

impl Eq for TimestampMode

impl StructuralPartialEq for TimestampMode