Skip to content

Add AMI dataset with MT-ASR annotations #87

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
EduardoPach merged 3 commits into from
Dec 30, 2025
Merged

Add AMI dataset with MT-ASR annotations #87

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
91cdfac
new: ami dataset aliases for mt-asr
EduardoPach Dec 29, 2025
422252f
new: ami dataset with diarization annotations
EduardoPach Dec 30, 2025
2ffe401
refactor: Undo callhome-english
EduardoPach Dec 30, 2025
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
35 changes: 32 additions & 3 deletions src/openbench/dataset/dataset_aliases.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -165,14 +165,43 @@ def register_dataset_aliases() -> None:
########## TRANSCRIPTION ##########

DatasetRegistry.register_alias(
"callhome-english",
DatasetConfig(
dataset_id=os.getenv("CALLHOME_ENGLISH_DATASET_REPO_ID", "argmaxinc/callhome-english"), split="test"
"ami-ihm-openbench",
DatasetConfig(dataset_id="argmaxinc/ami-openbench", split="test", subset="ihm-mix"),
supported_pipeline_types={
PipelineType.TRANSCRIPTION,
PipelineType.ORCHESTRATION,
PipelineType.DIARIZATION,
},
description=(
"AMI-IHM dataset for transcription, orchestration and diarization evaluation. "
"The audio files and the MT-ASR annotations were taken from https://github.com/BUTSpeechFIT/mt-asr-data-prep and processed to create the dataset. "
"The diarization annotations were taken from https://github.com/nttcslab-sp/diar-forced-alignment which should contain tighter annotations "
"when compared to using the ASR segments as the ground truth. See `Can We Really Repurpose Multi-Speaker ASR Corpus for Speaker Diarization?` for more details."
),
)

DatasetRegistry.register_alias(
"ami-sdm-openbench",
DatasetConfig(dataset_id="argmaxinc/ami-openbench", split="test", subset="sdm"),
supported_pipeline_types={
PipelineType.TRANSCRIPTION,
PipelineType.ORCHESTRATION,
PipelineType.DIARIZATION,
},
description=(
"AMI-SDM dataset for transcription, orchestration and diarization evaluation. "
"The audio files and the MT-ASR annotations were taken from https://github.com/BUTSpeechFIT/mt-asr-data-prep and processed to create the dataset. "
"The diarization annotations were taken from https://github.com/nttcslab-sp/diar-forced-alignment which should contain tighter annotations "
"when compared to using the ASR segments as the ground truth. See `Can We Really Repurpose Multi-Speaker ASR Corpus for Speaker Diarization?` for more details."
),
)

DatasetRegistry.register_alias(
"callhome-english",
DatasetConfig(
dataset_id=os.getenv("CALLHOME_ENGLISH_DATASET_REPO_ID", "argmaxinc/callhome-english"), split="test"
),
supported_pipeline_types={PipelineType.TRANSCRIPTION, PipelineType.ORCHESTRATION},
description=(
"Callhome English dataset for transcription and orchestration evaluation. "
"To use this dataset you need to buy the license for the audio files at https://catalog.ldc.upenn.edu/LDC97S42 and the license for the transcript files at https://catalog.ldc.upenn.edu/LDC97T14"
Expand Down
Loading