Spaces:
Runtime error
Runtime error
Commit
•
172ec24
1
Parent(s):
9e35e59
tidy
Browse files
app.py
CHANGED
@@ -1,11 +1,12 @@
|
|
1 |
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
|
2 |
from transformers.utils import is_flash_attn_2_available
|
|
|
3 |
import torch
|
4 |
import gradio as gr
|
5 |
import time
|
6 |
-
import os
|
7 |
|
8 |
BATCH_SIZE = 16
|
|
|
9 |
|
10 |
device = "cuda:0" if torch.cuda.is_available() else "cpu"
|
11 |
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
|
@@ -15,10 +16,11 @@ model = AutoModelForSpeechSeq2Seq.from_pretrained(
|
|
15 |
"openai/whisper-large-v2", torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True, use_flash_attention_2=use_flash_attention_2
|
16 |
)
|
17 |
distilled_model = AutoModelForSpeechSeq2Seq.from_pretrained(
|
18 |
-
"distil-whisper/distil-large-v2", torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True, use_flash_attention_2=use_flash_attention_2
|
19 |
)
|
20 |
|
21 |
if not use_flash_attention_2:
|
|
|
22 |
model = model.to_bettertransformer()
|
23 |
distilled_model = distilled_model.to_bettertransformer()
|
24 |
|
@@ -49,6 +51,7 @@ distil_pipe = pipeline(
|
|
49 |
chunk_length_s=15,
|
50 |
torch_dtype=torch_dtype,
|
51 |
device=device,
|
|
|
52 |
)
|
53 |
distil_pipe_forward = distil_pipe._forward
|
54 |
|
@@ -56,6 +59,20 @@ def transcribe(inputs):
|
|
56 |
if inputs is None:
|
57 |
raise gr.Error("No audio file submitted! Please record or upload an audio file before submitting your request.")
|
58 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
59 |
def _forward_distil_time(*args, **kwargs):
|
60 |
global distil_runtime
|
61 |
start_time = time.time()
|
@@ -92,7 +109,7 @@ if __name__ == "__main__":
|
|
92 |
"
|
93 |
>
|
94 |
<h1 style="font-weight: 900; margin-bottom: 7px; line-height: normal;">
|
95 |
-
|
96 |
</h1>
|
97 |
</div>
|
98 |
</div>
|
@@ -100,8 +117,11 @@ if __name__ == "__main__":
|
|
100 |
)
|
101 |
gr.HTML(
|
102 |
f"""
|
103 |
-
This demo
|
104 |
-
|
|
|
|
|
|
|
105 |
"""
|
106 |
)
|
107 |
audio = gr.components.Audio(type="filepath", label="Audio input")
|
@@ -117,4 +137,4 @@ if __name__ == "__main__":
|
|
117 |
inputs=audio,
|
118 |
outputs=[distil_transcription, distil_runtime, transcription, runtime],
|
119 |
)
|
120 |
-
demo.queue().launch()
|
|
|
1 |
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
|
2 |
from transformers.utils import is_flash_attn_2_available
|
3 |
+
from transformers.pipelines.audio_utils import ffmpeg_read
|
4 |
import torch
|
5 |
import gradio as gr
|
6 |
import time
|
|
|
7 |
|
8 |
BATCH_SIZE = 16
|
9 |
+
MAX_AUDIO_MINS = 30 # maximum audio input in minutes
|
10 |
|
11 |
device = "cuda:0" if torch.cuda.is_available() else "cpu"
|
12 |
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
|
|
|
16 |
"openai/whisper-large-v2", torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True, use_flash_attention_2=use_flash_attention_2
|
17 |
)
|
18 |
distilled_model = AutoModelForSpeechSeq2Seq.from_pretrained(
|
19 |
+
"distil-whisper/distil-large-v2", torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True, use_flash_attention_2=use_flash_attention_2
|
20 |
)
|
21 |
|
22 |
if not use_flash_attention_2:
|
23 |
+
# use flash attention from pytorch sdpa
|
24 |
model = model.to_bettertransformer()
|
25 |
distilled_model = distilled_model.to_bettertransformer()
|
26 |
|
|
|
51 |
chunk_length_s=15,
|
52 |
torch_dtype=torch_dtype,
|
53 |
device=device,
|
54 |
+
generate_kwargs={"language": "en", "task": "transcribe"},
|
55 |
)
|
56 |
distil_pipe_forward = distil_pipe._forward
|
57 |
|
|
|
59 |
if inputs is None:
|
60 |
raise gr.Error("No audio file submitted! Please record or upload an audio file before submitting your request.")
|
61 |
|
62 |
+
with open(inputs, "rb") as f:
|
63 |
+
inputs = f.read()
|
64 |
+
|
65 |
+
inputs = ffmpeg_read(inputs, pipe.feature_extractor.sampling_rate)
|
66 |
+
audio_length_mins = len(inputs) / pipe.feature_extractor.sampling_rate / 60
|
67 |
+
|
68 |
+
if audio_length_mins > MAX_AUDIO_MINS:
|
69 |
+
raise gr.Error(
|
70 |
+
f"To ensure fair usage of the Space, the maximum audio length permitted is {MAX_AUDIO_MINS} minutes."
|
71 |
+
f"Got an audio of length {round(audio_length_mins, 3)} minutes."
|
72 |
+
)
|
73 |
+
|
74 |
+
inputs = {"array": inputs, "sampling_rate": pipe.feature_extractor.sampling_rate}
|
75 |
+
|
76 |
def _forward_distil_time(*args, **kwargs):
|
77 |
global distil_runtime
|
78 |
start_time = time.time()
|
|
|
109 |
"
|
110 |
>
|
111 |
<h1 style="font-weight: 900; margin-bottom: 7px; line-height: normal;">
|
112 |
+
Whisper vs Distil-Whisper
|
113 |
</h1>
|
114 |
</div>
|
115 |
</div>
|
|
|
117 |
)
|
118 |
gr.HTML(
|
119 |
f"""
|
120 |
+
This demo shows a speed comparison between <a href="https://huggingface.co/openai/whisper-large-v2"> Whisper </a>
|
121 |
+
and <a href="https://huggingface.co/distil-whisper/distil-large-v2"> Distil-Whisper </a> for the same audio
|
122 |
+
file input. Both models use the <a href="https://huggingface.co/distil-whisper/distil-large-v2#long-form-transcription"> chunked long-form transcription algorithm </a>
|
123 |
+
in 🤗 Transformers with Flash Attention support. To ensure fair usage of the Space, we ask that audio
|
124 |
+
file inputs are kept to < 30 mins.
|
125 |
"""
|
126 |
)
|
127 |
audio = gr.components.Audio(type="filepath", label="Audio input")
|
|
|
137 |
inputs=audio,
|
138 |
outputs=[distil_transcription, distil_runtime, transcription, runtime],
|
139 |
)
|
140 |
+
demo.queue(max_size=10).launch()
|