> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/freerdp/freerdp/llms.txt
> Use this file to discover all available pages before exploring further.

# Audio

> Configure sound output and microphone input redirection in FreeRDP, including codecs and audio backends.

## Overview

FreeRDP supports redirecting audio in both directions:

* **Sound output** (`/sound`) — plays audio from the remote session on the local client.
* **Microphone input** (`/microphone`) — captures audio on the client and sends it to the remote session.

Both features are implemented as virtual channels (`rdpsnd` for output, `audin` for input) and are negotiated with the server at connect time.

***

## Build requirements

### Audio backends

FreeRDP supports the following platform audio backends. At least one must be available.

| Backend    | CMake flag                               | Platform          |
| ---------- | ---------------------------------------- | ----------------- |
| ALSA       | `-DWITH_ALSA=ON` (default)               | Linux             |
| PulseAudio | `-DWITH_PULSE=ON` (default)              | Linux             |
| OSS        | `-DWITH_OSS=ON` (default)                | Linux / BSD       |
| sndio      | Available when sndio headers are present | OpenBSD / FreeBSD |
| CoreAudio  | Built-in                                 | macOS             |
| WinMM      | Built-in                                 | Windows           |

### Audio codecs

The most widely supported codec is uncompressed PCM, which requires no additional libraries. Compressed codecs improve quality and reduce bandwidth.

| Codec           | Build flag                           | Notes                                                            |
| --------------- | ------------------------------------ | ---------------------------------------------------------------- |
| PCM             | Always available                     | Uncompressed, zero configuration                                 |
| AAC             | `-DWITH_FAAC=ON` / `-DWITH_FAAD2=ON` | Best quality; AAC encoding requires `-DWITH_DSP_EXPERIMENTAL=ON` |
| GSM             | `-DWITH_GSM=ON`                      | Older low-bandwidth codec                                        |
| FFmpeg codecs   | `-DWITH_DSP_FFMPEG=ON`               | Enables FFmpeg-based encode/decode pipeline                      |
| SOXR resampling | `-DWITH_SOXR=ON`                     | Optional resampling library                                      |

<Note>
  AAC is the preferred compressed codec for Windows 8 and newer servers. To enable AAC encoding
  (client-to-server microphone), you must also pass `-DWITH_DSP_EXPERIMENTAL=ON` at build time.
</Note>

Example build with PulseAudio and FFmpeg:

```bash theme={null}
cmake -GNinja \
  -DCMAKE_BUILD_TYPE=Release \
  -DWITH_PULSE=ON \
  -DWITH_DSP_FFMPEG=ON \
  -B build -S .
cmake --build build --target install
```

***

## Sound output

Enable sound redirection with the `/sound` flag.

```bash theme={null}
# Enable sound with default backend
xfreerdp /sound /v:rdp.example.com

# Specify the ALSA backend explicitly
xfreerdp /sound:sys:alsa /v:rdp.example.com

# Specify OSS with a device and format
xfreerdp /sound:sys:oss,dev:1,format:1 /v:rdp.example.com

# Use PulseAudio
xfreerdp /sound:sys:pulse /v:rdp.example.com
```

### Sound sub-options

| Sub-option      | Example    | Description                                       |
| --------------- | ---------- | ------------------------------------------------- |
| `sys:<backend>` | `sys:alsa` | Select the audio backend (`alsa`, `pulse`, `oss`) |
| `dev:<device>`  | `dev:1`    | Select a specific device index                    |
| `format:<n>`    | `format:1` | Select a specific audio format index              |

***

## Microphone input

Enable microphone redirection with the `/microphone` flag.

```bash theme={null}
# Enable microphone with default backend
xfreerdp /microphone /v:rdp.example.com

# Specify ALSA
xfreerdp /microphone:sys:alsa /v:rdp.example.com

# Specify OSS with a device
xfreerdp /microphone:sys:oss,dev:1,format:1 /v:rdp.example.com
```

The `/microphone` option accepts the same `sys:`, `dev:`, and `format:` sub-options as `/sound`.

***

## Using both simultaneously

```bash theme={null}
xfreerdp /sound:sys:pulse /microphone:sys:pulse /v:rdp.example.com
```

***

## Troubleshooting

<AccordionGroup>
  <Accordion title="No audio heard after connecting">
    * Confirm the build includes an audio backend: check for `rdpsnd-client-pulse.so` or `rdpsnd-client-alsa.so` in the plugin directory.
    * On Linux with PulseAudio, ensure the PulseAudio daemon is running: `pulseaudio --check`.
    * On Linux with ALSA, verify the device is not muted: `alsamixer`.
    * Enable debug logging to trace the codec negotiation:

      ```bash theme={null}
      WLOG_LEVEL=DEBUG xfreerdp /sound /v:rdp.example.com 2>&1 | grep -i rdpsnd
      ```
  </Accordion>

  <Accordion title="Microphone not working">
    * Ensure the build includes `audin` channel support with the correct backend.
    * On Linux, check that the input device is not muted and is set as the default capture device.
    * The server must support the Audio Input virtual channel (`audin`). Verify it is not blocked by group policy on the server.
  </Accordion>

  <Accordion title="Poor audio quality or high latency">
    * PCM is uncompressed and uses the most bandwidth. If bandwidth is limited, ensure a compressed codec (AAC or GSM) was negotiated. Check the debug log for the negotiated format.
    * Rebuild with `-DWITH_DSP_FFMPEG=ON` or `-DWITH_FAAD2=ON` to enable compressed codec support.
    * Consider enabling SOXR resampling (`-DWITH_SOXR=ON`) to improve sample rate conversion quality.
  </Accordion>

  <Accordion title="AAC codec not available">
    * AAC encoding requires `-DWITH_FAAC=ON` and `-DWITH_DSP_EXPERIMENTAL=ON`.
    * AAC decoding requires `-DWITH_FAAD2=ON` or `-DWITH_DSP_FFMPEG=ON`.
    * Rebuild with the appropriate flags and verify the codec appears in the negotiation log.
  </Accordion>
</AccordionGroup>