Skip to content

Fix Whisper doc + improve max_decode_ratio #1858

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
anautsch merged 9 commits into from
Feb 22, 2023
Merged

Fix Whisper doc + improve max_decode_ratio #1858

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
3050aa5
documentation
Adel-Moumen Feb 22, 2023
d9a65dc
doc seq2seq
Adel-Moumen Feb 22, 2023
ea3dbe5
fix max_decode_ratio
Adel-Moumen Feb 22, 2023
f866899
test only librispeech
Adel-Moumen Feb 22, 2023
1bf6cc1
extra requirements
Adel-Moumen Feb 22, 2023
2bbb01a
max_decode_ratio = 1
Adel-Moumen Feb 22, 2023
8dc20f0
precommit
Adel-Moumen Feb 22, 2023
c457492
fix overflow
Adel-Moumen Feb 22, 2023
c63f58e
remove print
Adel-Moumen Feb 22, 2023
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
Prev Previous commit
Next Next commit
test only librispeech
  • Loading branch information
@Adel-Moumen
Adel-Moumen committed Feb 22, 2023
commit f866899c005296925bf1bd8602ee3f0919cdb0cb
1 change: 1 addition & 0 deletions recipes/LibriSpeech/ASR/transformer/hparams/train_hf_whisper.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@ language: english
# Normalize the english inputs with
# the same normalization done in the paper
normalized_transcripts: True
test_only: False # Set it to True if you only want to do the evaluation

# Data files
data_folder: !PLACEHOLDER # e,g./path/to/LibriSpeech
Expand Down
18 changes: 10 additions & 8 deletions recipes/LibriSpeech/ASR/transformer/train_with_whisper.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -311,14 +311,16 @@ def text_pipeline(wrd):
asr_brain.tokenizer = tokenizer

# Training
asr_brain.fit(
asr_brain.hparams.epoch_counter,
train_data,
valid_data,
train_loader_kwargs=hparams["train_loader_kwargs"],
valid_loader_kwargs=hparams["valid_loader_kwargs"],
)

if hparams["test_only"] is False:
# Training
asr_brain.fit(
asr_brain.hparams.epoch_counter,
train_data,
valid_data,
train_loader_kwargs=hparams["train_loader_kwargs"],
valid_loader_kwargs=hparams["valid_loader_kwargs"],
)

# Testing
for k in test_datasets.keys(): # keys are test_clean, test_other etc
asr_brain.hparams.wer_file = os.path.join(
Expand Down