diff --git a/disagreement/audio.py b/disagreement/audio.py
index 9e58530..cd1eadf 100644
--- a/disagreement/audio.py
+++ b/disagreement/audio.py
@@ -5,6 +5,7 @@ from __future__ import annotations
 import asyncio
 import contextlib
 import io
+import shlex
 from typing import Optional, Union
 
 
@@ -35,15 +36,27 @@ class FFmpegAudioSource(AudioSource):
         A filename, URL, or file-like object to read from.
     """
 
-    def __init__(self, source: Union[str, io.BufferedIOBase]):
+    def __init__(
+        self,
+        source: Union[str, io.BufferedIOBase],
+        *,
+        before_options: Optional[str] = None,
+        options: Optional[str] = None,
+        volume: float = 1.0,
+    ):
         self.source = source
+        self.before_options = before_options
+        self.options = options
+        self.volume = volume
         self.process: Optional[asyncio.subprocess.Process] = None
         self._feeder: Optional[asyncio.Task] = None
 
     async def _spawn(self) -> None:
         if isinstance(self.source, str):
-            args = [
-                "ffmpeg",
+            args = ["ffmpeg"]
+            if self.before_options:
+                args += shlex.split(self.before_options)
+            args += [
                 "-i",
                 self.source,
                 "-f",
@@ -54,14 +67,18 @@ class FFmpegAudioSource(AudioSource):
                 "2",
                 "pipe:1",
             ]
+            if self.options:
+                args += shlex.split(self.options)
             self.process = await asyncio.create_subprocess_exec(
                 *args,
                 stdout=asyncio.subprocess.PIPE,
                 stderr=asyncio.subprocess.DEVNULL,
             )
         else:
-            args = [
-                "ffmpeg",
+            args = ["ffmpeg"]
+            if self.before_options:
+                args += shlex.split(self.before_options)
+            args += [
                 "-i",
                 "pipe:0",
                 "-f",
@@ -72,6 +89,8 @@ class FFmpegAudioSource(AudioSource):
                 "2",
                 "pipe:1",
             ]
+            if self.options:
+                args += shlex.split(self.options)
             self.process = await asyncio.create_subprocess_exec(
                 *args,
                 stdin=asyncio.subprocess.PIPE,
@@ -115,6 +134,7 @@ class FFmpegAudioSource(AudioSource):
             with contextlib.suppress(Exception):
                 self.source.close()
 
+
 class AudioSink:
     """Abstract base class for audio sinks."""
 
diff --git a/disagreement/voice_client.py b/disagreement/voice_client.py
index c771869..8696665 100644
--- a/disagreement/voice_client.py
+++ b/disagreement/voice_client.py
@@ -7,9 +7,26 @@ import asyncio
 import contextlib
 import socket
 import threading
+from array import array
+
+
+def _apply_volume(data: bytes, volume: float) -> bytes:
+    samples = array("h")
+    samples.frombytes(data)
+    for i, sample in enumerate(samples):
+        scaled = int(sample * volume)
+        if scaled > 32767:
+            scaled = 32767
+        elif scaled < -32768:
+            scaled = -32768
+        samples[i] = scaled
+    return samples.tobytes()
+
+
 from typing import TYPE_CHECKING, Optional, Sequence
 
 import aiohttp
+
 # The following import is correct, but may be flagged by Pylance if the virtual
 # environment is not configured correctly.
 from nacl.secret import SecretBox
@@ -180,6 +197,9 @@ class VoiceClient:
                 data = await self._current_source.read()
                 if not data:
                     break
+                volume = getattr(self._current_source, "volume", 1.0)
+                if volume != 1.0:
+                    data = _apply_volume(data, volume)
                 await self.send_audio_frame(data)
         finally:
             await self._current_source.close()
diff --git a/tests/test_voice_client.py b/tests/test_voice_client.py
index da052bc..1c6cb98 100644
--- a/tests/test_voice_client.py
+++ b/tests/test_voice_client.py
@@ -1,8 +1,11 @@
 import asyncio
+import io
+from array import array
 import pytest
 
+from disagreement.audio import AudioSource, FFmpegAudioSource
+
 from disagreement.voice_client import VoiceClient
-from disagreement.audio import AudioSource
 from disagreement.client import Client
 
 
@@ -137,3 +140,68 @@ async def test_play_and_switch_sources():
     await vc.play(DummySource([b"c"]))
 
     assert udp.sent == [b"a", b"b", b"c"]
+
+
+@pytest.mark.asyncio
+async def test_ffmpeg_source_custom_options(monkeypatch):
+    captured = {}
+
+    class DummyProcess:
+        def __init__(self):
+            self.stdout = io.BytesIO(b"")
+
+        async def wait(self):
+            return 0
+
+    async def fake_exec(*args, **kwargs):
+        captured["args"] = args
+        return DummyProcess()
+
+    monkeypatch.setattr(asyncio, "create_subprocess_exec", fake_exec)
+    src = FFmpegAudioSource(
+        "file.mp3", before_options="-reconnect 1", options="-vn", volume=0.5
+    )
+
+    await src._spawn()
+
+    cmd = captured["args"]
+    assert "-reconnect" in cmd
+    assert "-vn" in cmd
+    assert src.volume == 0.5
+
+
+@pytest.mark.asyncio
+async def test_voice_client_volume_scaling(monkeypatch):
+    ws = DummyWebSocket(
+        [
+            {"d": {"heartbeat_interval": 50}},
+            {"d": {"ssrc": 1, "ip": "127.0.0.1", "port": 4000}},
+            {"d": {"secret_key": []}},
+        ]
+    )
+    udp = DummyUDP()
+    vc = VoiceClient(
+        client=DummyVoiceClient(),
+        endpoint="ws://localhost",
+        session_id="sess",
+        token="tok",
+        guild_id=1,
+        user_id=2,
+        ws=ws,
+        udp=udp,
+    )
+    await vc.connect()
+    vc._heartbeat_task.cancel()
+
+    chunk = b"\x10\x00\x10\x00"
+    src = DummySource([chunk])
+    src.volume = 0.5
+
+    await vc.play(src)
+
+    samples = array("h")
+    samples.frombytes(chunk)
+    samples[0] = int(samples[0] * 0.5)
+    samples[1] = int(samples[1] * 0.5)
+    expected = samples.tobytes()
+    assert udp.sent == [expected]