mirror of
https://github.com/2noise/ChatTTS.git
synced 2026-09-01 14:55:37 +08:00
源文雨
37 lines
825 B
Python
37 lines
825 B
Python
from io import BufferedWriter, BytesIO
|
|
from typing import Dict
|
|
|
|
import av
|
|
|
|
|
|
video_format_dict: Dict[str, str] = {
|
|
"m4a": "mp4",
|
|
}
|
|
|
|
audio_format_dict: Dict[str, str] = {
|
|
"ogg": "libvorbis",
|
|
"mp4": "aac",
|
|
}
|
|
|
|
|
|
def wav2(i: BytesIO, o: BufferedWriter, format: str):
|
|
"""
|
|
https://github.com/fumiama/Retrieval-based-Voice-Conversion-WebUI/blob/412a9950a1e371a018c381d1bfb8579c4b0de329/infer/lib/audio.py#L20
|
|
"""
|
|
inp = av.open(i, "r")
|
|
format = video_format_dict.get(format, format)
|
|
out = av.open(o, "w", format=format)
|
|
format = audio_format_dict.get(format, format)
|
|
|
|
ostream = out.add_stream(format)
|
|
|
|
for frame in inp.decode(audio=0):
|
|
for p in ostream.encode(frame):
|
|
out.mux(p)
|
|
|
|
for p in ostream.encode(None):
|
|
out.mux(p)
|
|
|
|
out.close()
|
|
inp.close()
|