Skip to content

tweak whisper defaults for better long form audio decoding #45

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cjpais merged 2 commits into from
Jun 25, 2026
Merged

tweak whisper defaults for better long form audio decoding #45

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
df27945
tweak whisper default for better long form audio decoding
cjpais Jun 25, 2026
1942412
clean up comments
cjpais Jun 25, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions bindings/python/src/transcribe_cpp/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -975,7 +975,7 @@ def _resolve_family(self, family: "FamilyExtension", slot: str):
def run(self, pcm: PCMLike, *, task: Task = "transcribe",
language: str | None = None,
target_language: str | None = None,
timestamps: Timestamps = "none",
timestamps: Timestamps = "auto",
keep_special_tags: bool = False,
spec_k_drafts: int = -1,
family: FamilyExtension | None = None) -> Result:
Expand Down Expand Up @@ -1011,7 +1011,7 @@ def run(self, pcm: PCMLike, *, task: Task = "transcribe",
def run_batch(self, pcms: Sequence[PCMLike], *, task: Task = "transcribe",
language: str | None = None,
target_language: str | None = None,
timestamps: Timestamps = "none",
timestamps: Timestamps = "auto",
keep_special_tags: bool = False,
spec_k_drafts: int = -1,
family: FamilyExtension | None = None,
Expand Down Expand Up @@ -1339,7 +1339,7 @@ def transcribe(
task: Task = "transcribe",
language: str | None = None,
target_language: str | None = None,
timestamps: Timestamps = "none",
timestamps: Timestamps = "auto",
keep_special_tags: bool = False,
spec_k_drafts: int = -1,
family: FamilyExtension | None = None,
Expand Down
2 changes: 1 addition & 1 deletion bindings/rust/transcribe-cpp/src/session.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ impl Default for RunOptions {
fn default() -> Self {
RunOptions {
task: Task::Transcribe,
timestamps: TimestampKind::None,
timestamps: TimestampKind::Auto,
pnc: Pnc::Default,
itn: Itn::Default,
language: None,
Expand Down
4 changes: 2 additions & 2 deletions bindings/rust/transcribe-cpp/src/types.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,10 +29,10 @@ impl Task {
/// Requested (or returned) timestamp granularity.
#[derive(Debug, Clone, Copy, PartialEq, Eq, Default)]
pub enum TimestampKind {
/// Text only, no alignment data. The default for a run.
#[default]
/// Text only, no alignment data.
None,
/// "Richest the model supports" — resolved per-family at run time.
#[default]
Auto,
/// Segment-level start/end times.
Segment,
Expand Down
5 changes: 4 additions & 1 deletion bindings/swift/Sources/TranscribeCpp/Options.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -126,7 +126,10 @@ public struct RunOptions: Sendable {

public init(
task: TranscriptionTask = .transcribe,
timestamps: TimestampKind = .none,
// .auto ("richest the model supports", per-family) mirrors the C
// transcribe_run_params_init default: whisper resolves it to .segment
// (its robust path), no-timestamp families resolve to .none.
timestamps: TimestampKind = .auto,
pnc: Pnc = .default,
itn: Itn = .default,
language: String? = nil,
Expand Down
Loading
Loading