Comments (10)
marcwebbie
commented on August 25, 2024
What was the bitrate of the original file?
from pydub.
jiaaro
commented on August 25, 2024
He didn't say but since the resulting file was only a couple hundred bytes different in size (< 1% change), I assume it was also 128k
from pydub.
jiaaro
commented on August 25, 2024
Perhaps we should explicitly pass the duration meta data to ffmpeg when it's an mp3 file? It feels kludgy to me especially since we'll probably have to end up adding special cases like this for lots of formats if we go this route.
Maybe this is an ffmpeg bug that we should just pass upstream?
from pydub.
marcwebbie
commented on August 25, 2024
I think that we should try reproducing this bug before doing anything.
Yes, it may be a ffmpeg issue. It may also be a matter of args passed to
ffmpeg by pydub. I will try reproducing with some mp3 and using ffprobe to
check.
Le 29 oct. 2013 12:09, "James Robert" [email protected] a écrit :
Perhaps we should explicitly pass the duration meta data to ffmpeg when
it's an mp3 file? It feels kludgy to me especially since we'll probably
have to end up adding special cases like this for lots of formats if we go
this route.Maybe this is an ffmpeg bug that we should just pass upstream?
—
Reply to this email directly or view it on GitHub.
from pydub.
marcwebbie
commented on August 25, 2024
Debugging turned out into some funny bitrate converting. See ffprobe output, the bitrate in the "format" dict.
it shows "bit_rate": "128009" for song.mp3 and "bit_rate": "128019" for song_with_bitrate.mp3.
I agree with you, it looks like a ffmpeg issue.
Code to test de issue:
# issue28.py
import shlex
import subprocess
import os
from pydub import AudioSegment
def test_pydub():
print("original_song: godfather.mp3")
original_song = 'godfather.mp3'
print("Building 'song' AudioSegment with from_mp3(original_song)...")
song = AudioSegment.from_mp3(original_song)
print("Building 'song2' AudioSegment with from_file(original_song)...")
song2 = AudioSegment.from_file(original_song)
print("Exporting 'song' to mp3 as 'song.mp3'...")
song.export("song.mp3", format="mp3")
print("Exporting 'song' to mp3 with arg bitrate=128 as 'song_with_bitrate.mp3'...")
song.export("song_with_bitrate.mp3", format="mp3", bitrate="128k")
print("Exporting 'song2' to mp3 as 'song2.mp3'...")
song2.export("song2.mp3", format="mp3")
print("Exporting 'song2' to mp3 with arg bitrate=128 as 'song2_with_bitrate.mp3'...")
song2.export("song2_with_bitrate.mp3", format="mp3", bitrate="128k")
print("Getting exported files size...")
original_song_size = os.path.getsize(original_song)
song_size = os.path.getsize("song.mp3")
song_with_bitrate_size = os.path.getsize("song_with_bitrate.mp3")
song2_size = os.path.getsize("song2.mp3")
song2_with_bitrate_size = os.path.getsize("song2_with_bitrate.mp3")
print("Size of original_song: {}".format(original_song_size))
print("Size of 'song.mp3': {}".format(song_size))
print("size of 'song_with_bitrate.mp3': {}".format(song_with_bitrate_size))
print("size of 'song2.mp3': {}".format(song2_size))
print("size of 'song2_with_bitrate.mp3': {}".format(song2_with_bitrate_size))
def test_with_ffprobe():
print("")
print("ffprobe song.mp3")
print("==========================================================")
subprocess.call(
shlex.split("ffprobe -v quiet -print_format json -show_format -show_streams song.mp3"))
print("")
print("ffprobe song_with_bitrate.mp3")
print("==========================================================")
subprocess.call(
shlex.split("ffprobe -v quiet -print_format json -show_format -show_streams song_with_bitrate.mp3"))
if __name__ == "__main__":
test_pydub()
test_with_ffprobe()The output confirms the issue:
[marc@arch pydub]$ python issue28.py
original_song: godfather.mp3
Building 'song' AudioSegment with from_mp3(original_song)...
Building 'song2' AudioSegment with from_file(original_song)...
Exporting 'song' to mp3 as 'song.mp3'...
ffmpeg call: ['ffmpeg', '-y', '-f', 'wav', '-i', '/tmp/tmp3egck9', '-f', 'mp3', '/tmp/tmp1f647m']
Exporting 'song' to mp3 with arg bitrate=128 as 'song_with_bitrate.mp3'...
ffmpeg call: ['ffmpeg', '-y', '-f', 'wav', '-i', '/tmp/tmp7f51ep', '-b:a', '128k', '-f', 'mp3', '/tmp/tmpt56ya2']
Exporting 'song2' to mp3 as 'song2.mp3'...
ffmpeg call: ['ffmpeg', '-y', '-f', 'wav', '-i', '/tmp/tmpecx3iy', '-f', 'mp3', '/tmp/tmpcgar9y']
Exporting 'song2' to mp3 with arg bitrate=128 as 'song2_with_bitrate.mp3'...
ffmpeg call: ['ffmpeg', '-y', '-f', 'wav', '-i', '/tmp/tmp0pa7bt', '-b:a', '128k', '-f', 'mp3', '/tmp/tmp8rrgnc']
Getting exported files size...
Size of original_song: 2988022
Size of 'song.mp3': 2988635
size of 'song_with_bitrate.mp3': 2988870
size of 'song2.mp3': 2988635
size of 'song2_with_bitrate.mp3': 2988870
ffprobe song.mp3
==========================================================
{
"streams": [
{
"index": 0,
"codec_name": "mp3",
"codec_long_name": "MP3 (MPEG audio layer 3)",
"codec_type": "audio",
"codec_time_base": "1/44100",
"codec_tag_string": "[0][0][0][0]",
"codec_tag": "0x0000",
"sample_fmt": "s16p",
"sample_rate": "44100",
"channels": 2,
"bits_per_sample": 0,
"r_frame_rate": "0/0",
"avg_frame_rate": "0/0",
"time_base": "1/14112000",
"start_pts": 0,
"start_time": "0.000000",
"duration_ts": 2635776000,
"duration": "186.775510",
"bit_rate": "128000",
"disposition": {
"default": 0,
"dub": 0,
"original": 0,
"comment": 0,
"lyrics": 0,
"karaoke": 0,
"forced": 0,
"hearing_impaired": 0,
"visual_impaired": 0,
"clean_effects": 0,
"attached_pic": 0
}
}
],
"format": {
"filename": "song.mp3",
"nb_streams": 1,
"format_name": "mp3",
"format_long_name": "MP2/3 (MPEG audio layer 2/3)",
"start_time": "0.000000",
"duration": "186.775510",
"size": "2988635",
"bit_rate": "128009",
"tags": {
"encoder": "Lavf55.12.100"
}
}
}
ffprobe song_with_bitrate.mp3
==========================================================
{
"streams": [
{
"index": 0,
"codec_name": "mp3",
"codec_long_name": "MP3 (MPEG audio layer 3)",
"codec_type": "audio",
"codec_time_base": "1/44100",
"codec_tag_string": "[0][0][0][0]",
"codec_tag": "0x0000",
"sample_fmt": "s16p",
"sample_rate": "44100",
"channels": 2,
"bits_per_sample": 0,
"r_frame_rate": "0/0",
"avg_frame_rate": "0/0",
"time_base": "1/14112000",
"start_pts": 0,
"start_time": "0.000000",
"duration_ts": 2635776000,
"duration": "186.775510",
"bit_rate": "128000",
"disposition": {
"default": 0,
"dub": 0,
"original": 0,
"comment": 0,
"lyrics": 0,
"karaoke": 0,
"forced": 0,
"hearing_impaired": 0,
"visual_impaired": 0,
"clean_effects": 0,
"attached_pic": 0
}
}
],
"format": {
"filename": "song_with_bitrate.mp3",
"nb_streams": 1,
"format_name": "mp3",
"format_long_name": "MP2/3 (MPEG audio layer 2/3)",
"start_time": "0.000000",
"duration": "186.775510",
"size": "2988870",
"bit_rate": "128019",
"tags": {
"encoder": "Lavf55.12.100"
}
}
}from pydub.
jiaaro
commented on August 25, 2024
Nice detective work! I have mixed feelings about doing nothing though.
What do you think?
from pydub.
marcwebbie
commented on August 25, 2024
I've found some flaws on pydub due to that issue, it's more complicated than we've thought, keep the issue open. I'm working on something to lessen this issue.
from pydub.
lepinsk
commented on August 25, 2024
Yikes; sorry, didn't mean for my issue to show up here. Sorry about that...
from pydub.
jpasden
commented on August 25, 2024
I'm running into this same issue, and the discrepancies are larger. Like 12s VS 14s.
It's bad enough that I have to output to WAV (no problems there) and use other software to convert to MP3.
(@jiaaro I'm the guy that posted to StackOverflow the other day about adding silence. Love pyDub so much! If I can find a workaround for this, it will be just perfect...)
from pydub.
jiaaro
commented on August 25, 2024
@jpasden So far this appears to be an issue with ffmpeg - you can try upgrading ffmpeg, or using the libav fork instead, which has a better reputation for fixing bugs.
edit: Other people around the internet seems to be seeing this issue as well. Some people seem to be having success with MP3Diags and mp3val, but the first thing I'd try is using the latest libav instead of ffmpeg.
from pydub.
Related Issues (20)
- M4A Chapters
- Pydub import error
- Improve memory footprint of make_chunks
- Calling AudioSegment.export() with an integer bitrate causes Python subprocess exception HOT 1
- is there a faster way to play audio?
- using base64 audio data on AudioSegment
- test failures on python 3.12 HOT 2
- Tell
- Create new tag version.
- Disable pydub RuntimeWarning if not applicable to deployment HOT 1
- How do you guys install ffmpeg or libav inside a python-image? HOT 1
- Unwanted Logs!!
- Failed to decode audio in `.mkv` files with `flac` codec [analysis and solution attached]
- Chunked Audio not Playing
- "SyntaxWarning: invalid escape sequence \("
- pyinstaller compatibility?
- Inconsistency with audio segment duration on silence or Sine
- MP3 exports unreadable/corrupted HOT 1
- app crash after playing a boing sound with exit error code -1073741819 (0xC0000005)
- pyaudioop.py def ratecv doesn't return something
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pydub.