kavlab
commited on
Commit
·
8a3847d
1
Parent(s):
5a8691c
added kuznetsov_av/api.py
Browse files- kuznetsov_av/api.py +31 -0
kuznetsov_av/api.py
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from fastapi import FastAPI
|
2 |
+
from pydantic import BaseModel
|
3 |
+
import text_to_speech_converter as t2s
|
4 |
+
import base64
|
5 |
+
|
6 |
+
class Request(BaseModel):
|
7 |
+
"""
|
8 |
+
Input text.
|
9 |
+
"""
|
10 |
+
text: str
|
11 |
+
|
12 |
+
class Response(BaseModel):
|
13 |
+
"""
|
14 |
+
Result of text-to-audio generation.
|
15 |
+
audio - base64 string
|
16 |
+
"""
|
17 |
+
audio: str
|
18 |
+
sampling_rate: int
|
19 |
+
|
20 |
+
|
21 |
+
app = FastAPI()
|
22 |
+
|
23 |
+
@app.post('/text-to-speach/')
|
24 |
+
def text_to_speach(entity: Request) -> Response:
|
25 |
+
"""
|
26 |
+
Text-to-audio generation method using text_to_speech_converter.
|
27 |
+
"""
|
28 |
+
synthesiser = t2s.load_model()
|
29 |
+
embeddings_dataset = t2s.load_speaker_dataset()
|
30 |
+
audio, sampling_rate = t2s.text_to_speech(entity.text, synthesiser, embeddings_dataset)
|
31 |
+
return Response(audio=base64.b32encode(audio), sampling_rate=sampling_rate)
|