Hello Nauman, and vr14:
I’m in a similar situation; I believe I’ve followed the setup instructions diligently, but I’m also getting the same error when running the experiment. I’ll try to enumerate my relevant state below:
OS: Ubuntu 20.04, with CUDA installed and verified working.
AMI Dataset lives at: demarco/datasets/ami, with a tree structure of:
(py38) demarco@demarco-ubuntu:~/dev/datasets/ami$ tree . --filelimit 27
.
├── corpus
│ └── ES2002a
│ ├── audio
│ │ └── ES2002a.Mix-Headset.wav
│ └── video
│ ├── ES2002a.Closeup1.avi
│ ├── ES2002a.Closeup2.avi
│ ├── ES2002a.Closeup3.avi
│ ├── ES2002a.Closeup4.avi
│ ├── ES2002a.Corner.avi
│ └── ES2002a.Overhead.avi
└── manual_annot
├── 00README_MANUAL.txt
├── abstractive [142 entries exceeds filelimit, not opening dir]
├── AMI-metadata.xml
├── argumentation
│ ├── ae [376 entries exceeds filelimit, not opening dir]
│ ├── ar [94 entries exceeds filelimit, not opening dir]
│ └── dis [95 entries exceeds filelimit, not opening dir]
├── configuration
│ ├── amiConfig.xml
│ └── amiSubConfig.xml
├── corpusdoc [73 entries exceeds filelimit, not opening dir]
├── corpusResources
│ ├── meetings.xml
│ └── participants.xml
├── decision
│ └── manual [47 entries exceeds filelimit, not opening dir]
├── dialogueActs [695 entries exceeds filelimit, not opening dir]
├── disfluency [160 entries exceeds filelimit, not opening dir]
├── extractive [274 entries exceeds filelimit, not opening dir]
├── focus [56 entries exceeds filelimit, not opening dir]
├── handGesture [61 entries exceeds filelimit, not opening dir]
├── headGesture [173 entries exceeds filelimit, not opening dir]
├── LICENCE.txt
├── manifest_1.7.html
├── MANIFEST_MANUAL.txt
├── movement [498 entries exceeds filelimit, not opening dir]
├── namedEntities [468 entries exceeds filelimit, not opening dir]
├── ontologies
│ ├── ae-types.xml
│ ├── ap-types.xml
│ ├── ar-types.xml
│ ├── chunks.xml
│ ├── chunk-types.xml
│ ├── da-types.xml
│ ├── default-topics.xml
│ ├── dis-types.xml
│ ├── dsfl-types.xml
│ ├── floor-types.xml
│ ├── foa-targets.xml
│ ├── leg-targets.xml
│ ├── ne-types.xml
│ ├── rse-types.xml
│ ├── rsr-types.xml
│ ├── subj-types.xml
│ └── you-types.xml
├── participantRoles
│ ├── ES2002d.A.role.xml
│ ├── ES2002d.B.role.xml
│ ├── ES2002d.C.role.xml
│ ├── ES2002d.D.role.xml
│ ├── ES2008b.A.role.xml
│ ├── ES2008b.B.role.xml
│ ├── ES2008b.C.role.xml
│ ├── ES2008b.D.role.xml
│ ├── ES2008d.A.role.xml
│ ├── ES2008d.B.role.xml
│ ├── ES2008d.C.role.xml
│ ├── ES2008d.D.role.xml
│ ├── ES2009d.A.role.xml
│ ├── ES2009d.B.role.xml
│ ├── ES2009d.C.role.xml
│ ├── ES2009d.D.role.xml
│ ├── IS1003d.A.role.xml
│ ├── IS1003d.B.role.xml
│ ├── IS1003d.C.role.xml
│ └── IS1003d.D.role.xml
├── participantSummaries [323 entries exceeds filelimit, not opening dir]
├── resource.xml
├── segments [687 entries exceeds filelimit, not opening dir]
├── topics [139 entries exceeds filelimit, not opening dir]
├── words [687 entries exceeds filelimit, not opening dir]
└── youUsages [63 entries exceeds filelimit, not opening dir]
30 directories, 54 files
Note I simply downloaded the Edinburgh ES 2002a meeting, selected “Low-size DivX videos” and “Headset mix”.
I installed extra_requirements.txt and updated hparams/ecapa_tdnn.yaml as follows:
seed: 1234
__set_seed: !apply:torch.manual_seed [!ref <seed>]
# Folders
# data: http://groups.inf.ed.ac.uk/ami/download/
data_folder: /home/demarco/dev/datasets/ami/corpus/
manual_annot_folder: /home/demarco/dev/datasets/ami/manual_annot/
output_folder: results/ami/sd_ecapa_tdnn
save_folder: !ref <output_folder>/save
device: 'cuda:0'
I only changed the data_folder and manual_annot_folder paths, but also removed a trailing / from output_folder; I noted under a debugger that full_ref_rttm_file had a double // in its path; unclear if that could negatively affect things.
My error output is identical:
(py38) demarco@demarco-ubuntu:~/dev/when/ml/speechbrain/recipes/AMI/Diarization$ python experiment.py hparams/ecapa_tdnn.yaml
speechbrain.core - Beginning experiment!
speechbrain.core - Experiment folder: results/ami/sd_ecapa_tdnn
speechbrain.pretrained.fetching - Fetch embedding_model.ckpt: Using existing file/symlink in results/ami/sd_ecapa_tdnn/save/embedding_model.ckpt.
speechbrain.utils.parameter_transfer - Loading pretrained files for: embedding_model
__main__ - Tuning for nn (Multiple iterations over AMI Dev set)
__main__ - Diarizing dev set
0it [00:00, ?it/s]
speechbrain.core - Exception:
Traceback (most recent call last):
File "experiment.py", line 443, in <module>
best_nn = dev_nn_tuner(full_csv, "dev")
File "experiment.py", line 292, in dev_nn_tuner
concate_rttm_file = diarize_dataset(
File "experiment.py", line 229, in diarize_dataset
concate_rttm_file = out_rttm_dir + "/sys_output.rttm"
UnboundLocalError: local variable 'out_rttm_dir' referenced before assignment
I’m not sure how to proceed. Any help you can offer would be greatly appreciated. Thank you in advance for your help and your contributions to this excellent library.