Update app.py
Browse files
app.py
CHANGED
|
@@ -9,11 +9,35 @@ from pathlib import Path
|
|
| 9 |
from pydub import AudioSegment
|
| 10 |
|
| 11 |
def get_silence(duration_ms=1000):
|
| 12 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
|
| 14 |
# Get all available voices
|
| 15 |
async def get_voices():
|
| 16 |
-
|
|
|
|
| 17 |
|
| 18 |
async def text_to_speech_segment(text_segment, voice, rate, pitch):
|
| 19 |
"""Processes a single text segment for voice commands and generates audio."""
|
|
|
|
| 9 |
from pydub import AudioSegment
|
| 10 |
|
| 11 |
def get_silence(duration_ms=1000):
|
| 12 |
+
# Create silent audio segment with specified parameters
|
| 13 |
+
silent_audio = AudioSegment.silent(
|
| 14 |
+
duration=duration_ms,
|
| 15 |
+
frame_rate=24000 # 24kHz sampling rate
|
| 16 |
+
)
|
| 17 |
+
|
| 18 |
+
# Set audio parameters
|
| 19 |
+
silent_audio = silent_audio.set_channels(1) # Mono
|
| 20 |
+
silent_audio = silent_audio.set_sample_width(4) # 32-bit (4 bytes per sample)
|
| 21 |
+
|
| 22 |
+
with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
|
| 23 |
+
# Export with specific bitrate and codec parameters
|
| 24 |
+
silent_audio.export(
|
| 25 |
+
tmp_file.name,
|
| 26 |
+
format="mp3",
|
| 27 |
+
bitrate="48k",
|
| 28 |
+
parameters=[
|
| 29 |
+
"-ac", "1", # Mono
|
| 30 |
+
"-ar", "24000", # Sample rate
|
| 31 |
+
"-sample_fmt", "s32", # 32-bit samples
|
| 32 |
+
"-codec:a", "libmp3lame" # MP3 codec
|
| 33 |
+
]
|
| 34 |
+
)
|
| 35 |
+
return tmp_file.name
|
| 36 |
|
| 37 |
# Get all available voices
|
| 38 |
async def get_voices():
|
| 39 |
+
voices = await edge_tts.list_voices()
|
| 40 |
+
return {f"{v['ShortName']} - {v['Locale']} ({v['Gender']})": v['ShortName'] for v in voices}
|
| 41 |
|
| 42 |
async def text_to_speech_segment(text_segment, voice, rate, pitch):
|
| 43 |
"""Processes a single text segment for voice commands and generates audio."""
|