Spaces:
Sleeping
Sleeping
Commit
•
72c65b6
1
Parent(s):
93acf16
Add code
Browse files- app.py +12 -0
- requirements.txt +2 -1
app.py
CHANGED
@@ -14,12 +14,21 @@ from pydub import AudioSegment
|
|
14 |
from transformers import AutoTokenizer, AutoFeatureExtractor, set_seed
|
15 |
from transformers.generation.streamers import BaseStreamer
|
16 |
from huggingface_hub import InferenceClient
|
|
|
|
|
|
|
17 |
|
18 |
device = "cuda:0" if torch.cuda.is_available() else "mps" if torch.backends.mps.is_available() else "cpu"
|
19 |
torch_dtype = torch.float16 if device != "cpu" else torch.float32
|
20 |
|
21 |
repo_id = "parler-tts/parler_tts_mini_v0.1"
|
22 |
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
model = ParlerTTSForConditionalGeneration.from_pretrained(
|
24 |
repo_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True
|
25 |
).to(device)
|
@@ -205,6 +214,9 @@ def generate_base(subject, setting, ):
|
|
205 |
gr.Info("Story Generated", duration=3)
|
206 |
story = response.choices[0].message.content
|
207 |
|
|
|
|
|
|
|
208 |
play_steps_in_s = 4.0
|
209 |
play_steps = int(frame_rate * play_steps_in_s)
|
210 |
streamer = ParlerTTSStreamer(model, device=device, play_steps=play_steps)
|
|
|
14 |
from transformers import AutoTokenizer, AutoFeatureExtractor, set_seed
|
15 |
from transformers.generation.streamers import BaseStreamer
|
16 |
from huggingface_hub import InferenceClient
|
17 |
+
import nltk
|
18 |
+
nltk.download('punkt')
|
19 |
+
|
20 |
|
21 |
device = "cuda:0" if torch.cuda.is_available() else "mps" if torch.backends.mps.is_available() else "cpu"
|
22 |
torch_dtype = torch.float16 if device != "cpu" else torch.float32
|
23 |
|
24 |
repo_id = "parler-tts/parler_tts_mini_v0.1"
|
25 |
|
26 |
+
jenny_repo_id = "ylacombe/parler-tts-mini-jenny-30H"
|
27 |
+
|
28 |
+
model = ParlerTTSForConditionalGeneration.from_pretrained(
|
29 |
+
jenny_repo_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True
|
30 |
+
).to(device)
|
31 |
+
|
32 |
model = ParlerTTSForConditionalGeneration.from_pretrained(
|
33 |
repo_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True
|
34 |
).to(device)
|
|
|
214 |
gr.Info("Story Generated", duration=3)
|
215 |
story = response.choices[0].message.content
|
216 |
|
217 |
+
model_input = story.replace("\n", " ").strip()
|
218 |
+
model_input = nltk.sent_tokenize(model_input)
|
219 |
+
|
220 |
play_steps_in_s = 4.0
|
221 |
play_steps = int(frame_rate * play_steps_in_s)
|
222 |
streamer = ParlerTTSStreamer(model, device=device, play_steps=play_steps)
|
requirements.txt
CHANGED
@@ -1,3 +1,4 @@
|
|
1 |
https://gradio-builds.s3.amazonaws.com/bed454c3d22cfacedc047eb3b0ba987b485ac3fd/gradio-4.40.0-py3-none-any.whl
|
2 |
git+https://github.com/huggingface/parler-tts.git
|
3 |
-
accelerate
|
|
|
|
1 |
https://gradio-builds.s3.amazonaws.com/bed454c3d22cfacedc047eb3b0ba987b485ac3fd/gradio-4.40.0-py3-none-any.whl
|
2 |
git+https://github.com/huggingface/parler-tts.git
|
3 |
+
accelerate
|
4 |
+
nltk
|