Remove pydub dependency (#529)

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-authored-by: gradio-pr-bot <gradio-pr-bot@users.noreply.github.com>

Author: 3 people

.changeset/solid-sides-look.md

@@ -0,0 +1,5 @@
+---
+"trackio": patch
+---
+
+feat:Remove pydub dependency

examples/crash-and-resume-same-run-name.py

@@ -22,15 +22,8 @@
 import argparse
 import math
 import uuid
-import warnings
 
-warnings.filterwarnings(
-    "ignore",
-    category=SyntaxWarning,
-    module=r"pydub\.utils",
-)
-
-import trackio  # noqa: E402
+import trackio
 
 DEFAULT_PROJECT = f"crash-and-resume-demo-{uuid.uuid4().hex[:8]}"
 DEFAULT_RUN_NAME = "trainer-job-42"

pyproject.toml

@@ -21,7 +21,6 @@ dependencies = [
     "numpy<3.0.0",
     "pillow<13.0.0",
     "orjson>=3.0,<4.0.0",
-    "pydub<1.0.0",
     "tomli>=2.0.0; python_version < '3.11'",
 ]
 classifiers = [

tests/unit/test_audio_writer.py

@@ -1,5 +1,7 @@
+import json
 import math
 import shutil
+import subprocess
 import wave
 from pathlib import Path
 
@@ -34,6 +36,29 @@ def _has_ffmpeg() -> bool:
     return shutil.which("ffmpeg") is not None
 
 
+def _has_ffprobe() -> bool:
+    return shutil.which("ffprobe") is not None
+
+
+def _probe_audio(path: Path) -> tuple[int, int]:
+    out = subprocess.check_output(
+        [
+            "ffprobe",
+            "-v",
+            "error",
+            "-select_streams",
+            "a:0",
+            "-show_entries",
+            "stream=sample_rate,channels",
+            "-of",
+            "json",
+            str(path),
+        ]
+    )
+    stream = json.loads(out)["streams"][0]
+    return int(stream["sample_rate"]), int(stream["channels"])
+
+
 @pytest.mark.parametrize("channels", [1, 2])
 def test_write_wav_mono_and_stereo_with_float_normalization(
     tmp_path: Path, channels: int
@@ -56,11 +81,11 @@ def test_write_wav_mono_and_stereo_with_float_normalization(
     assert 32000 <= max_abs <= 32767
 
 
-@pytest.mark.skipif(not _has_ffmpeg(), reason="ffmpeg not available")
+@pytest.mark.skipif(
+    not (_has_ffmpeg() and _has_ffprobe()), reason="ffmpeg/ffprobe not available"
+)
 @pytest.mark.parametrize("channels", [1, 2])
 def test_write_mp3_mono_and_stereo(tmp_path: Path, channels: int) -> None:
-    from pydub import AudioSegment
-
     mono = _tone(0.1, 440.0, SAMPLE_RATE, amp=0.5)
     data = mono if channels == 1 else np.stack([mono, mono], axis=1)
 
@@ -69,9 +94,10 @@ def test_write_mp3_mono_and_stereo(tmp_path: Path, channels: int) -> None:
         data=data, sample_rate=SAMPLE_RATE, filename=out, format="mp3"
     )
 
-    seg = AudioSegment.from_file(str(out), format="mp3")
-    assert seg.frame_rate == SAMPLE_RATE
-    assert seg.channels == channels
+    assert out.exists() and out.stat().st_size > 0
+    sr, ch = _probe_audio(out)
+    assert sr == SAMPLE_RATE
+    assert ch == channels
 
 
 @pytest.mark.parametrize(

trackio/media/audio.py

@@ -1,11 +1,12 @@
 import os
 import shutil
+import subprocess
 import warnings
+import wave
 from pathlib import Path
 from typing import Literal
 
 import numpy as np
-from pydub import AudioSegment
 
 from trackio.media.media import TrackioMedia
 from trackio.media.utils import check_ffmpeg_installed, check_path
@@ -156,12 +157,33 @@ def write_audio(
             check_ffmpeg_installed()
 
         channels = 1 if pcm.ndim == 1 else pcm.shape[1]
-        audio = AudioSegment(
-            pcm.tobytes(),
-            frame_rate=sample_rate,
-            sample_width=2,  # int16
-            channels=channels,
-        )
-
-        file = audio.export(str(filename), format=format)
-        file.close()
+        pcm_bytes = pcm.tobytes()
+
+        if format == "wav":
+            with wave.open(str(filename), "wb") as wf:
+                wf.setnchannels(channels)
+                wf.setsampwidth(2)
+                wf.setframerate(sample_rate)
+                wf.writeframes(pcm_bytes)
+        else:
+            cmd = [
+                "ffmpeg",
+                "-y",
+                "-loglevel",
+                "error",
+                "-f",
+                "s16le",
+                "-ar",
+                str(sample_rate),
+                "-ac",
+                str(channels),
+                "-i",
+                "pipe:0",
+                str(filename),
+            ]
+            proc = subprocess.run(
+                cmd, input=pcm_bytes, capture_output=True, check=False
+            )
+            if proc.returncode != 0:
+                stderr = proc.stderr.decode("utf-8", errors="replace").strip()
+                raise RuntimeError(f"ffmpeg failed to encode {format} audio: {stderr}")

trackio/media/utils.py

@@ -20,7 +20,7 @@ def check_ffmpeg_installed() -> None:
     """Raise an error if ffmpeg is not available on the system PATH."""
     if shutil.which("ffmpeg") is None:
         raise RuntimeError(
-            "ffmpeg is required to write video but was not found on your system. "
+            "ffmpeg is required to write this media format but was not found on your system. "
             "Please install ffmpeg and ensure it is available on your PATH."
         )