sangeet2020
commited on
Commit
•
f551734
1
Parent(s):
1675342
modify yaml file; seems everything is working
Browse files
asr.yaml
CHANGED
@@ -6,17 +6,17 @@
|
|
6 |
# Authors: Sangeet Sagar 2023
|
7 |
# ################################
|
8 |
|
9 |
-
# URL for the biggest whisper model.
|
10 |
# URL for the biggest Fairseq english whisper model.
|
11 |
whisper_hub: openai/whisper-large-v2
|
12 |
language: german
|
13 |
|
|
|
|
|
14 |
normalized_transcripts: true
|
|
|
15 |
|
16 |
-
|
17 |
sample_rate: 16000
|
18 |
-
freeze_whisper: True
|
19 |
-
freeze_encoder: True
|
20 |
|
21 |
# These values are only used for the searchers.
|
22 |
# They needs to be hardcoded and should not be changed with Whisper.
|
@@ -27,27 +27,38 @@ timestamp_index: 50363
|
|
27 |
eos_index: 50257
|
28 |
bos_index: 50258
|
29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
|
31 |
-
|
32 |
-
whisper: &id003 !new:speechbrain.lobes.models.huggingface_whisper.HuggingFaceWhisper
|
33 |
source: !ref <whisper_hub>
|
34 |
freeze: !ref <freeze_whisper>
|
35 |
freeze_encoder: !ref <freeze_encoder>
|
36 |
save_path: whisper_checkpoints
|
37 |
encoder_only: False
|
38 |
|
39 |
-
decoder: &id006 !new:speechbrain.decoders.seq2seq.S2SWhisperGreedySearch
|
40 |
-
model: *id003
|
41 |
-
bos_index: 50363
|
42 |
-
eos_index: 50257
|
43 |
-
min_decode_ratio: 0.0
|
44 |
-
max_decode_ratio: 1.0
|
45 |
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
|
|
|
|
51 |
|
52 |
modules:
|
53 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
# Authors: Sangeet Sagar 2023
|
7 |
# ################################
|
8 |
|
|
|
9 |
# URL for the biggest Fairseq english whisper model.
|
10 |
whisper_hub: openai/whisper-large-v2
|
11 |
language: german
|
12 |
|
13 |
+
# Normalize the english inputs with
|
14 |
+
# the same normalization done in the paper
|
15 |
normalized_transcripts: true
|
16 |
+
test_only: false # Set it to True if you only want to do the evaluation
|
17 |
|
18 |
+
auto_mix_prec: False
|
19 |
sample_rate: 16000
|
|
|
|
|
20 |
|
21 |
# These values are only used for the searchers.
|
22 |
# They needs to be hardcoded and should not be changed with Whisper.
|
|
|
27 |
eos_index: 50257
|
28 |
bos_index: 50258
|
29 |
|
30 |
+
# Decoding parameters
|
31 |
+
min_decode_ratio: 0.0
|
32 |
+
max_decode_ratio: 0.1
|
33 |
+
test_beam_size: 8
|
34 |
+
|
35 |
+
# Model parameters
|
36 |
+
freeze_whisper: True
|
37 |
+
freeze_encoder: True
|
38 |
+
|
39 |
+
|
40 |
|
41 |
+
whisper: !new:speechbrain.lobes.models.huggingface_whisper.HuggingFaceWhisper
|
|
|
42 |
source: !ref <whisper_hub>
|
43 |
freeze: !ref <freeze_whisper>
|
44 |
freeze_encoder: !ref <freeze_encoder>
|
45 |
save_path: whisper_checkpoints
|
46 |
encoder_only: False
|
47 |
|
|
|
|
|
|
|
|
|
|
|
|
|
48 |
|
49 |
+
|
50 |
+
decoder: !new:speechbrain.decoders.seq2seq.S2SWhisperGreedySearch
|
51 |
+
model: !ref <whisper>
|
52 |
+
bos_index: !ref <timestamp_index>
|
53 |
+
eos_index: !ref <eos_index>
|
54 |
+
min_decode_ratio: !ref <min_decode_ratio>
|
55 |
+
max_decode_ratio: !ref <max_decode_ratio>
|
56 |
|
57 |
modules:
|
58 |
+
whisper: !ref <whisper>
|
59 |
+
decoder: !ref <decoder>
|
60 |
+
|
61 |
+
|
62 |
+
pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
|
63 |
+
loadables:
|
64 |
+
whisper: !ref <whisper>
|