Skip to content

fix: pad audio arrays to same shape if sample rates differ #1880

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

Open
chrisammon3000 wants to merge 3 commits into
base: main
Choose a base branch
from
Open

fix: pad audio arrays to same shape if sample rates differ #1880

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
7cc0cc7
fix: pad audio arrays to same shape if sample rates differ
chrisammon3000 Apr 3, 2024
a574002
Merge branch 'main' into fix-audio-tensor-mismatch
chrisammon3000 Apr 3, 2024
5f3141c
Merge branch 'main' into fix-audio-tensor-mismatch
Jun 4, 2024
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
Next Next commit
fix: pad audio arrays to same shape if sample rates differ
  • Loading branch information
@chrisammon3000
chrisammon3000 committed Apr 3, 2024
commit 7cc0cc7c8f9777e2a47c04391cb2cd6f4e8dde78
18 changes: 18 additions & 0 deletions docarray/typing/bytes/video_bytes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,4 @@
import warnings
from io import BytesIO
from typing import TYPE_CHECKING, List, NamedTuple, TypeVar

Expand Down Expand Up @@ -80,6 +81,11 @@ class MyDoc(BaseDoc):

video_frames.append(frame.to_ndarray(format='rgb24'))

# Pad audio arrays to same shape if sample rates differ
if len({arr.shape for arr in audio_frames}) > 1:
warnings.warn('Audio frames have different sample rates')
audio_frames = self._pad_arrays_to_same_shape(audio_frames)

if len(audio_frames) == 0:
audio = parse_obj_as(AudioNdArray, np.array(audio_frames))
else:
Expand All @@ -89,3 +95,15 @@ class MyDoc(BaseDoc):
indices = parse_obj_as(NdArray, keyframe_indices)

return VideoLoadResult(video=video, audio=audio, key_frame_indices=indices)

@staticmethod
def _pad_arrays_to_same_shape(arrays: List[np.ndarray]) -> List[np.ndarray]:
# Calculate the maximum number of samples in any array
max_samples = max(arr.shape[1] for arr in arrays)

# Pad arrays with fewer samples
for i, arr in enumerate(arrays):
if arr.shape[1] < max_samples:
arrays[i] = np.pad(arr, ((0, 0), (0, max_samples - arr.shape[1])))

return arrays