From e0833da5dc7132006310761f7e166e120f516da3 Mon Sep 17 00:00:00 2001 From: Howard Date: Thu, 15 May 2025 14:30:55 +0800 Subject: [PATCH 1/2] Fix: Ensure integer tensor indexing in get_wildcard_emission() --- whisperx/alignment.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/whisperx/alignment.py b/whisperx/alignment.py index b77a7f1bf..34fbbbba3 100644 --- a/whisperx/alignment.py +++ b/whisperx/alignment.py @@ -424,7 +424,7 @@ def get_wildcard_emission(frame_emission, tokens, blank_id): wildcard_mask = (tokens == -1) # Get scores for non-wildcard positions - regular_scores = frame_emission[tokens.clamp(min=0)] # clamp to avoid -1 index + regular_scores = frame_emission[tokens.clamp(min=0).long()] # clamp to avoid -1 index # Create a mask and compute the maximum value without modifying frame_emission max_valid_score = frame_emission.clone() # Create a copy From 429658d4ccefa55244bcdccd5d179795436093e4 Mon Sep 17 00:00:00 2001 From: Barabazs <31799121+Barabazs@users.noreply.github.com> Date: Fri, 27 Jun 2025 07:18:39 +0000 Subject: [PATCH 2/2] chore: bump version to 3.4.2 --- pyproject.toml | 2 +- uv.lock | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/pyproject.toml b/pyproject.toml index bd22bfdad..9b849abae 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -2,7 +2,7 @@ urls = { repository = "https://github.com/m-bain/whisperx" } authors = [{ name = "Max Bain" }] name = "whisperx" -version = "3.4.1" +version = "3.4.2" description = "Time-Accurate Automatic Speech Recognition using Whisper." readme = "README.md" requires-python = ">=3.9, <3.13" diff --git a/uv.lock b/uv.lock index e24a7a4e8..7ed518b2c 100644 --- a/uv.lock +++ b/uv.lock @@ -2788,7 +2788,7 @@ wheels = [ [[package]] name = "whisperx" -version = "3.4.1" +version = "3.4.2" source = { editable = "." } dependencies = [ { name = "ctranslate2" },