Update tts_script.py
Browse files- tts_script.py +35 -53
tts_script.py
CHANGED
@@ -2,8 +2,8 @@ import time
|
|
2 |
import requests
|
3 |
import pathlib
|
4 |
from io import BytesIO
|
5 |
-
from flask import Flask, request,
|
6 |
-
from
|
7 |
|
8 |
# Flask App Setup
|
9 |
app = Flask(__name__)
|
@@ -11,11 +11,9 @@ app = Flask(__name__)
|
|
11 |
# ElevenLabs API Configuration
|
12 |
ELEVENLABS_API_URL = "https://api.elevenlabs.io/v1/text-to-speech"
|
13 |
HEADERS = {"User-Agent": "TTSApp"}
|
14 |
-
# Change cache directory to a writable location
|
15 |
CACHE_DIR = pathlib.Path("/tmp/audio_cache")
|
16 |
CACHE_DIR.mkdir(parents=True, exist_ok=True)
|
17 |
|
18 |
-
|
19 |
# Available Voices
|
20 |
ALL_VOICES = {
|
21 |
"Brian": "nPczCjzI2devNBz1zQrb",
|
@@ -35,65 +33,49 @@ def generate_audio(text, voice):
|
|
35 |
filename = CACHE_DIR / f"{int(time.time())}.mp3"
|
36 |
sentences = split_sentences(text)
|
37 |
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
part_number, audio_data = future.result()
|
57 |
-
if audio_data:
|
58 |
-
audio_chunks[part_number] = audio_data
|
59 |
-
|
60 |
-
combined_audio = BytesIO()
|
61 |
-
for part_number in sorted(audio_chunks.keys()):
|
62 |
-
combined_audio.write(audio_chunks[part_number])
|
63 |
-
|
64 |
-
with open(filename, "wb") as f:
|
65 |
-
f.write(combined_audio.getvalue())
|
66 |
-
|
67 |
return filename.as_posix()
|
68 |
|
69 |
# Flask Routes
|
70 |
-
@app.route("/")
|
71 |
def home():
|
72 |
-
|
73 |
-
|
74 |
-
|
75 |
-
|
76 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
77 |
<label>Voice:</label>
|
78 |
<select name="voice">
|
79 |
<option value="Brian">Brian</option>
|
80 |
<option value="Alice">Alice</option>
|
81 |
<option value="Will">Will</option>
|
82 |
-
</select>
|
83 |
-
<button type="submit">Generate</button>
|
84 |
</form>
|
85 |
-
|
86 |
-
|
87 |
-
@app.route("/tts", methods=["POST"])
|
88 |
-
def tts():
|
89 |
-
text = request.form.get("text")
|
90 |
-
voice = request.form.get("voice", "Brian")
|
91 |
-
|
92 |
-
if not text:
|
93 |
-
return jsonify({"error": "Text is required!"})
|
94 |
-
|
95 |
-
audio_file = generate_audio(text, voice)
|
96 |
-
return send_file(audio_file, as_attachment=True)
|
97 |
|
98 |
if __name__ == "__main__":
|
99 |
app.run(debug=True, host="0.0.0.0", port=5000)
|
|
|
2 |
import requests
|
3 |
import pathlib
|
4 |
from io import BytesIO
|
5 |
+
from flask import Flask, request, render_template, send_file
|
6 |
+
from pydub import AudioSegment
|
7 |
|
8 |
# Flask App Setup
|
9 |
app = Flask(__name__)
|
|
|
11 |
# ElevenLabs API Configuration
|
12 |
ELEVENLABS_API_URL = "https://api.elevenlabs.io/v1/text-to-speech"
|
13 |
HEADERS = {"User-Agent": "TTSApp"}
|
|
|
14 |
CACHE_DIR = pathlib.Path("/tmp/audio_cache")
|
15 |
CACHE_DIR.mkdir(parents=True, exist_ok=True)
|
16 |
|
|
|
17 |
# Available Voices
|
18 |
ALL_VOICES = {
|
19 |
"Brian": "nPczCjzI2devNBz1zQrb",
|
|
|
33 |
filename = CACHE_DIR / f"{int(time.time())}.mp3"
|
34 |
sentences = split_sentences(text)
|
35 |
|
36 |
+
audio_chunks = []
|
37 |
+
for sentence in sentences:
|
38 |
+
response = requests.post(
|
39 |
+
f"{ELEVENLABS_API_URL}/{ALL_VOICES[voice]}",
|
40 |
+
headers=HEADERS,
|
41 |
+
json={"text": sentence, "model_id": "eleven_multilingual_v2"},
|
42 |
+
timeout=20
|
43 |
+
)
|
44 |
+
if response.ok:
|
45 |
+
audio_chunks.append(BytesIO(response.content))
|
46 |
+
|
47 |
+
# Combine all audio parts
|
48 |
+
combined_audio = AudioSegment.empty()
|
49 |
+
for chunk in audio_chunks:
|
50 |
+
chunk_audio = AudioSegment.from_file(chunk, format="mp3")
|
51 |
+
combined_audio += chunk_audio
|
52 |
+
|
53 |
+
combined_audio.export(filename, format="mp3")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
54 |
return filename.as_posix()
|
55 |
|
56 |
# Flask Routes
|
57 |
+
@app.route("/", methods=["GET", "POST"])
|
58 |
def home():
|
59 |
+
if request.method == "POST":
|
60 |
+
text = request.form["text"]
|
61 |
+
voice = request.form.get("voice", "Brian")
|
62 |
+
audio_file = generate_audio(text, voice)
|
63 |
+
return send_file(audio_file, as_attachment=True)
|
64 |
+
|
65 |
+
return """
|
66 |
+
<h1>Text-to-Speech Generator</h1>
|
67 |
+
<form method="post">
|
68 |
+
<label>Text:</label><br>
|
69 |
+
<textarea name="text" rows="4" cols="50" required></textarea><br>
|
70 |
<label>Voice:</label>
|
71 |
<select name="voice">
|
72 |
<option value="Brian">Brian</option>
|
73 |
<option value="Alice">Alice</option>
|
74 |
<option value="Will">Will</option>
|
75 |
+
</select><br><br>
|
76 |
+
<button type="submit">Generate & Download</button>
|
77 |
</form>
|
78 |
+
"""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
79 |
|
80 |
if __name__ == "__main__":
|
81 |
app.run(debug=True, host="0.0.0.0", port=5000)
|