Upload folder using huggingface_hub
Browse files- tts_gradio.py +9 -7
tts_gradio.py
CHANGED
@@ -10,7 +10,7 @@ import os
|
|
10 |
import re
|
11 |
import tempfile
|
12 |
# import librosa
|
13 |
-
|
14 |
# import torch
|
15 |
# from torch import no_grad, LongTensor
|
16 |
# import commons
|
@@ -166,7 +166,8 @@ def openai(text, name):
|
|
166 |
# exactly as it was in the original request.
|
167 |
#data = '{\n "model": "tts-1",\n "input": "The quick brown fox jumped over the lazy dog.",\n "voice": "alloy"\n }'
|
168 |
#response = requests.post('https://api.openai.com/v1/audio/speech', headers=headers, data=data)
|
169 |
-
|
|
|
170 |
|
171 |
def elevenlabs(text,name):
|
172 |
url = f"https://api.elevenlabs.io/v1/text-to-speech/{eleven_id_model_name_dict[name]}"
|
@@ -221,11 +222,11 @@ def microsoft(text, name, style="Neural"):
|
|
221 |
data=data,
|
222 |
)
|
223 |
# breakpoint()
|
224 |
-
timestamp = int(time.time()*10000)
|
225 |
-
path = f'/tmp/output_{timestamp}.wav'
|
226 |
-
with open(path, 'wb') as f:
|
227 |
-
|
228 |
-
return "Success",
|
229 |
|
230 |
if __name__ == '__main__':
|
231 |
parser = argparse.ArgumentParser()
|
@@ -271,6 +272,7 @@ if __name__ == '__main__':
|
|
271 |
app.queue(max_size=10)
|
272 |
app.launch(share=True)
|
273 |
# _, audio = microsoft(all_example,microsoft_model_list[0])
|
|
|
274 |
# print(audio)
|
275 |
# with open("test99.mp3", "wb") as f:
|
276 |
# f.write(audio.content)
|
|
|
10 |
import re
|
11 |
import tempfile
|
12 |
# import librosa
|
13 |
+
import numpy as np
|
14 |
# import torch
|
15 |
# from torch import no_grad, LongTensor
|
16 |
# import commons
|
|
|
166 |
# exactly as it was in the original request.
|
167 |
#data = '{\n "model": "tts-1",\n "input": "The quick brown fox jumped over the lazy dog.",\n "voice": "alloy"\n }'
|
168 |
#response = requests.post('https://api.openai.com/v1/audio/speech', headers=headers, data=data)
|
169 |
+
out_arr = np.frombuffer(response.content, dtype=np.uint8)
|
170 |
+
return "Success", (24000,out_arr)
|
171 |
|
172 |
def elevenlabs(text,name):
|
173 |
url = f"https://api.elevenlabs.io/v1/text-to-speech/{eleven_id_model_name_dict[name]}"
|
|
|
222 |
data=data,
|
223 |
)
|
224 |
# breakpoint()
|
225 |
+
# timestamp = int(time.time()*10000)
|
226 |
+
# path = f'/tmp/output_{timestamp}.wav' # TODO: disk might full.
|
227 |
+
# with open(path, 'wb') as f:
|
228 |
+
# f.write(response.content)
|
229 |
+
return "Success", response
|
230 |
|
231 |
if __name__ == '__main__':
|
232 |
parser = argparse.ArgumentParser()
|
|
|
272 |
app.queue(max_size=10)
|
273 |
app.launch(share=True)
|
274 |
# _, audio = microsoft(all_example,microsoft_model_list[0])
|
275 |
+
# breakpoint()
|
276 |
# print(audio)
|
277 |
# with open("test99.mp3", "wb") as f:
|
278 |
# f.write(audio.content)
|