Skip to content

Matt/parakeet runner timestamps #1

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

Draft
mattjcly wants to merge 9 commits into
base: main
Choose a base branch
from
Draft

Matt/parakeet runner timestamps #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
d3bab3a
Initial parakeet timestamps impl
mattjcly Jan 8, 2026
29c30a8
dev annotation
mattjcly Jan 8, 2026
565e78c
Subword/token timestamps
mattjcly Jan 8, 2026
9d072c5
Cleanup comments
mattjcly Jan 8, 2026
4c5402b
format
mattjcly Jan 8, 2026
64e61a8
Choose type of timestamps you want
mattjcly Jan 8, 2026
f322904
cleanups
mattjcly Jan 9, 2026
9d31012
no clang format
mattjcly Jan 9, 2026
6744199
More cleanup
mattjcly Jan 9, 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
0 .clang-format → examples/models/parakeet/.clang-format
View file
Open in desktop
File renamed without changes.
1 change: 1 addition & 0 deletions examples/models/parakeet/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -71,3 +71,4 @@ From the executorch root directory:
| `--audio_path` | Path to input audio file (.wav) |
| `--tokenizer_path` | Path to tokenizer file (default: `tokenizer.json`) |
| `--data_path` | Path to data file (.ptd) for delegate data (optional, required for CUDA) |
| `--timestamps` | Timestamp output mode: none\|subword\|word\|segment\|all (requires `window_stride` + `encoder_subsampling_factor` in model metadata) |
4 changes: 4 additions & 0 deletions examples/models/parakeet/export_parakeet_tdt.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -351,13 +351,17 @@ def export_all(model):
)

sample_rate = model.preprocessor._cfg.sample_rate
window_stride = float(model.preprocessor._cfg.window_stride)
encoder_subsampling_factor = int(getattr(model.encoder, "subsampling_factor", 8))
metadata = {
"num_rnn_layers": num_layers,
"pred_hidden": pred_hidden,
"joint_hidden": joint_hidden,
"vocab_size": model.tokenizer.vocab_size,
"blank_id": model.tokenizer.vocab_size,
"sample_rate": sample_rate,
"window_stride": window_stride,
"encoder_subsampling_factor": encoder_subsampling_factor,
}

return programs, metadata
Expand Down
Loading