Adding ffmpeg formatting

.gitignore

@@ -1,2 +1,12 @@
 app.log
 .DS_Store
+__pycache__/commons.cpython-312.pyc
+__pycache__/constant.cpython-312.pyc
+models/__pycache__/attentions.cpython-312.pyc
+models/__pycache__/models.cpython-312.pyc
+models/__pycache__/modules.cpython-312.pyc
+models/__pycache__/transforms.cpython-312.pyc
+models/__pycache__/utils.cpython-312.pyc
+monotonic_align/__pycache__/__init__.cpython-312.pyc
+text/__pycache__/__init__.cpython-312.pyc
+text/__pycache__/cleaners.cpython-312.pyc

app.py

@@ -1,7 +1,6 @@
 import torch
 from torch import LongTensor, no_grad
 import soundfile as sf
-import base64
 import commons
 from models import utils
 from io import BytesIO
@@ -12,8 +11,9 @@ from flask import Flask, request, jsonify, send_file
 import threading
 from constant import speakerList
 import logging
+import imageio_ffmpeg
 from pydub import AudioSegment
-from pydub.utils import which
+
 
 app = Flask(__name__)
 
@@ -51,12 +51,8 @@ umas_name_to_id = {name: int(idx) for idx, name in umas.items()}
 # Define a lock for thread-safe model inference
 model_lock = threading.Lock()
 
-# Explicitly Set FFmpeg Path for Pydub
+# Set FFmpeg Path for Pydub
-AudioSegment.converter = "/usr/bin/ffmpeg"
+AudioSegment.converter = imageio_ffmpeg.get_ffmpeg_exe()
-
-# Debug: Print the resolved path of ffmpeg
-print("FFmpeg path:", which("ffmpeg"))
-
 
 def get_text(text, hps):
     """

requirements.txt

@@ -5,25 +5,15 @@ Cython==3.0.11
 filelock==3.16.1
 Flask==3.1.0
 fsspec==2024.10.0
+imageio-ffmpeg==0.5.1
 itsdangerous==2.2.0
 Jinja2==3.1.4
 MarkupSafe==3.0.2
 mpmath==1.3.0
 networkx==3.4.2
 numpy==2.1.3
-nvidia-cublas-cu12==12.4.5.8
-nvidia-cuda-cupti-cu12==12.4.127
-nvidia-cuda-nvrtc-cu12==12.4.127
-nvidia-cuda-runtime-cu12==12.4.127
-nvidia-cudnn-cu12==9.1.0.70
-nvidia-cufft-cu12==11.2.1.3
-nvidia-curand-cu12==10.3.5.147
-nvidia-cusolver-cu12==11.6.1.9
-nvidia-cusparse-cu12==12.3.1.170
-nvidia-nccl-cu12==2.21.5
-nvidia-nvjitlink-cu12==12.4.127
-nvidia-nvtx-cu12==12.4.127
 pycparser==2.22
+pydub==0.25.1
 pyopenjtalk==0.3.4
 scipy==1.14.1
 setuptools==75.5.0
@@ -31,7 +21,6 @@ soundfile==0.12.1
 sympy==1.13.1
 torch==2.5.1
 tqdm==4.67.0
-triton==3.1.0
 typing_extensions==4.12.2
 Unidecode==1.3.8
 Werkzeug==3.1.3