The 2025 Google Summer of Code project for Sound Open Firmware (SOF) focused on a Virtual DAI: a software-based Digital Audio Interface intended to let developers run and debug SOF audio pipelines without connecting physical DAI hardware. The contributor’s final report describes work across SOF, Zephyr and NXP’s linux-imx stack, with i.MX8MP used as the project-specific hardware context. It also records ongoing upstreaming and data-copy problems, so the report does not establish that every goal was merged or that the result is production-ready.
What Sound Open Firmware is
Sound Open Firmware is an open-source project that provides audio DSP firmware and a software development kit. It is developer infrastructure, not a consumer audio device or a finished playback product. The public project materials describe a development flow involving the Zephyr SDK, QEMU, source repositories, Python environments and system dependencies. The project identifies its license as BSD Clause 3.
As an Amazon Associate I earn from qualifying purchases.
SOF operates in a host/DSP architecture. A host operating system communicates with DSP firmware through inter-process communication (IPC). Those messages support stream setup, pipeline creation, run, pause and stop commands, and parameter changes; the architecture documentation names IPC3 and IPC4 protocol generations.
What the 2025 GSoC project tried to build
The project’s central deliverable was a Virtual DAI component. Instead of requiring a physical codec or external DAI endpoint, the virtual interface supplies a software endpoint that an audio pipeline can use for development and testing.
#1 Best Overall
- Powerful Processor: Equipped with ESP32-S3R8 Xtensa 32-bit LX7 dual-core processor, up to 240MHz main frequency. Supports 2.4GHz Wi-Fi (802.11 b/g/n) and Bluetooth 5 (LE), with onboard antenna. Built-in 512KB of SRAM and 384KB ROM, with onboard 8MB PSRAM and an external 16MB Flash memory.
- Driver and Touch LCD: Onboard 1.83inch IPS Capacitive Touch Display, 240 × 284 resolution, 65K color. Built-in ST7789P display driver and CST816D capacitive touch chip, using SPI and I2C communication respectively, effectively saving the IO resources. Adopts Type-C port to improve user convenience and device compatibility.
- Supports Offline Speech recognition and AI Speech Interaction: Allows access to online large model platforms such as ChatGPT, DeepSeek, Doubao, etc. Onboard ES8311 audio codec chip and ES7210 echo cancellation circuit to meet daily audio application scenarios.
- Multifunctional Sensor: Onboard QMI8658 6-axis IMU (3-axis accelerometer and 3-axis gyroscope) for detecting motion gestures, counting steps, etc; PCF85063 RTC chip connected to the battry via the AXP2101 for uninterrupted power supply; Onboard PWR and BOOT programmable buttons for easy custom function development.
- Rich Peripheral Interface: Reserved 1 × I2C, 1 × UART and 1 × USB pads for external device connection and debugging, enabling flexible peripheral configuration. Onboard TF card slot for extended storage and fast data transfer, suitable for applications such as data recording and media playback, simplifying circuit design.
Intended uses
- Debugging pipeline behavior when the target DAI hardware is unavailable.
- Rapid prototyping of SOF audio graphs before hardware integration.
- A foundation for software loopback and memory-to-memory audio processing.
These goals describe the purpose of the component, not a claim that a general-purpose loopback product was completed. The final report presents software loopback as a future direction.
Goals listed in the contributor report
- Implement and upstream the SOF Virtual DAI component.
- Create a simple playback-and-record pipeline using NoCodec.
- Add NoCodec support for an i.MX8MP board.
- Support playback by discarding incoming data.
- Support capture by generating zero-filled or patterned frames and logging activity.
The report also notes a data-copy issue affecting some operations. That caveat matters when interpreting the project as an engineering work product rather than a finished feature.
Rank #2
- 🖥️ Professional HUB75 LED Matrix Controller: Designed as a LED matrix controller board, this module supports HUB75 RGB LED matrix panels for smart displays, animated signs, dashboards, and custom interface projects. Optimized for embedded display control and graphical applications with smooth performance.
- 🎤 Dual Microphone Audio Interaction Board: Built as an audio interaction development board, it features an onboard dual microphones array, ES7210 echo cancellation chip, and ES8311 codec chip for voice pickup, sound processing, and speaker output. Suitable for smart voice interfaces and multimedia display systems.
- 💾 High Performance Development Board: This ESP32-S3 development board integrates 32MB Flash, 16MB PSRAM, TF card slot, USB Type-C, UART, I2C, GPIO, and programmable buttons, giving developers flexible storage, debugging, and expansion options for advanced embedded projects.
- 🧭 Sensor Rich Smart Display Board: As a smart display board, it includes onboard 6-axis IMU motion sensor, temperature and humidity sensor, plus RTC clock chip for gesture sensing, environment monitoring, and real-time clock functions. Ideal for interactive dashboards and AIoT systems.
- ⚙️ LVGL GUI Development Platform: This LVGL development board supports ESP-IDF, Arduino, and LVGL GUI development, helping users quickly build custom user interfaces and scalable RGB matrix display systems. Dual power input design supports cascading panels for larger installations.
How the implementation was divided
The reported work crossed three layers of the embedded audio stack. The following summary reflects the contributor’s linked work items and does not imply that all changes were finally merged.
| Layer | Reported work | What it contributes |
|---|---|---|
| SOF firmware | Virtual DAI support for i.MX8MP, including topology and emul_dma integration |
Connects the virtual endpoint to an SOF pipeline and emulated DMA path. |
| Zephyr | Virtual DAI driver, device-tree binding, build/configuration changes and API tests | Provides the RTOS-side driver and interfaces used by the firmware environment. |
| linux-imx | Virtual DAI type/configuration and an i.MX8M backend | Adds host-kernel support for the virtual interface on the NXP Linux stack. |
| linux-imx NoCodec | NoCodec support | Allows a pipeline arrangement that does not depend on a physical audio codec. |
What “without physical DAI hardware” means
A conventional SOF test path often depends on a real DAI, codec and board wiring. A Virtual DAI substitutes a software endpoint so the pipeline can still receive stream-control messages and process audio buffers. For a basic playback path, the implementation can consume and discard data. For capture, it can produce zeros or a repeatable pattern, with logging to make activity visible.
Rank #3
- Built for Custom Integration: Keep control of the enclosure, mounting and final device layout. The open-board format fits robots, kiosks, custom voice devices and embedded prototypes where flexible mechanical integration matters.
- Onboard Voice Processing: XVF3800 performs AEC, beamforming, de-reverberation, DoA, VAD, AGC and noise suppression before audio reaches your application, helping reduce downstream audio preprocessing.
- 360° Far-Field Voice Capture: Four MEMS microphones in a circular array support speech pickup from different directions at distances up to 5 m, so users do not need to speak toward one fixed microphone position.
- XIAO ESP32S3 for Embedded Voice: The pre-soldered XIAO adds Wi-Fi, Bluetooth Low Energy and MCU-side control for connected voice interfaces, local wake-word projects and custom embedded applications.
- Firmware Options: Ships with Standard I2S firmware for XIAO ESP32S3 and is not a USB audio device by default; switch to USB firmware for host audio or use dedicated 48 kHz HA I2S firmware for Home Assistant and ESPHome Voice; configurations are separate.
This makes it possible to exercise parts of stream setup, pipeline instantiation and buffer handling before a physical audio device is available. It does not reproduce every electrical, clocking, codec or board-level behavior of real hardware. Tests that depend on those properties still require the corresponding physical platform.
Is i.MX8MP required?
No. i.MX8MP is the board and processor context described by this GSoC implementation, not a general SOF requirement. The report also refers to an i.MX8M Linux backend. A developer reproducing the exact reported path will need to verify the matching NXP board, kernel changes, Zephyr configuration, SOF firmware and topology; another SOF platform may require different support.
Rank #4
- High-performance ADAU1467 DSP Core Board designed for advanced processing applications.
- Supports a wide range of formats and provides exceptional sound quality for professional systems.
- Low power consumption design ensures efficient operation, making it ideal for embedded solutions.
- Versatile compatibility with various devices, enhancing your projects with ease.
- Compact and user-friendly design, perfect for engineers and developers looking to integrate DSP technology into their products.
The report identifies an “NXP i.MX 8M Plus evaluation kit” as relevant development hardware, but it does not specify an exact retail SKU or establish that a particular kit is mandatory. It also does not establish current marketplace availability.
What a developer would need to reproduce the work
- Choose the target platform. Decide whether you are reproducing the i.MX8MP-oriented path or adapting the concept to another supported SOF platform.
- Align the software versions. Match the host kernel, linux-imx changes, Zephyr build, SOF firmware, topology and IPC generation. SOF firmware packages are platform- and version-specific, so a binary should not be copied into an unrelated setup without checking release compatibility.
- Set up the SOF development environment. The project’s public setup flow references the Zephyr SDK, QEMU, source repositories, Python environments and required system packages.
- Build the virtual endpoint and NoCodec path. Enable the relevant driver, device-tree and build configuration, then use the reported topology and
emul_dmaintegration where applicable. - Exercise playback and capture separately. Playback should follow the discard-data behavior described in the report; capture should check generated zero or patterned frames and the associated logs.
- Investigate buffer-copy behavior. The contributor explicitly reported data-copy problems for some playback/capture operations, so successful compilation or stream creation should not be treated as proof that every transfer mode works.
What the final report establishes—and what it does not
Established by the report
- The project targeted a Virtual DAI for hardware-independent pipeline development.
- Work was reported in SOF, Zephyr and linux-imx components.
- NoCodec and i.MX8MP-specific work were among the stated goals.
- Playback discard behavior, generated capture frames and logging were part of the described design.
- Data-copy issues remained for parts of playback and capture.
Not established by the cited material
- That every pull request was merged into the relevant upstream projects.
- That the implementation passed an official GSoC evaluation.
- That it is production-ready.
- A quantitative performance benchmark, latency figure or measured development-time saving.
- The number of physical boards supported.
- A specific currently available retail board SKU.
The contributor’s report labels the SOF implementation as “currently upstreaming.” That wording should be preserved: linked pull requests show the scope of the work, but they are not proof of final merge or release status.
Best Value
- ESP32-S3R8 Processor--- Equipped with ESP32-S3R8 Xtensa 32-bit LX7 dual-core processor, up to 240MHz main frequency. Supports 2.4GHz W-i-F-i (802.11 b/g/n) and Blue--tooth 5 (LE), with onboard antenna. Built in 512KB of SRAM and 384KB ROM, with onboard 8MB PSRAM and an external 16MB Flash memory.
- AMOLED Touch Screen--- Onboard 1.8inch AMOLED display for clear color picture display, 368 x 448 resolution, 16.7M color, 178° wide viewing angle. Compared to those traditional LCD displays, the AMOLED screen features precise light-control capability, representing more delicate colors, more picture details, and more vivid video image.
- Onboard Audio Codec---Supports high-quality audio processing, providing clear and high-quality audio input and output. Supports Offline Speech recognition and AI Speech Interaction---Allows access to online large model platforms to support more AI application scenarios.
- For Various Smart Devices---Suitable For Various Smart Devices Development, Can Realize Human-Computer Interaction Function. Supports installing ba|tte|ry inside the case for independent operation. （Note: this version doesn't include ba|tte|ry ） Dedicated Black Case---with removable back cover for easy embedded into the projects and DIY design.
- Sensor and Chip---Onboard QMI8658 6-axis IMU (3-axis accelerometer and 3-axis gyroscope) for detecting motion gesture, counting steps, etc. Built-in SH8601 display driver and FT3168 capacitive touch chip, using QSPI and I2C communication respectively, effectively saving the IO resources.
Why this project matters to SOF development
Audio-pipeline debugging is often slowed by missing codecs, boards or DAI wiring. A virtual endpoint can let developers validate host-to-DSP control flow, pipeline construction and buffer processing earlier in the cycle. Because the reported changes span firmware, an RTOS driver and the host Linux side, the project also demonstrates that a seemingly small audio interface feature crosses multiple integration boundaries.
Its value is therefore mainly for developers and maintainers. It is not an end-user audio enhancement by itself, and the cited material provides no benchmark showing how much time or latency it saves.
How this fits the 2025 GSoC program
Google’s Work Product Submission Guidelines say that contributors should explain goals, accomplishments, current state, remaining tasks and what did or did not reach upstream. They also state: “The idea of GSoC isn’t that contributors churn out code — it’s important that the code be potentially useful to the hosting Open Source project.” The SOF report is best read through that lens: it documents a useful cross-stack direction while openly recording unfinished integration work.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteIn a March 24, 2025 Google Open Source Blog announcement, GSoC program administrators Stephanie Taylor, Mary Radomile and Lucila Ortíz wrote that “Since 2005, GSoC has successfully brought over 21,000 new contributors from 123 countries into the open source community.” That is a program-wide cumulative figure, not a count for SOF or this project. The 2025 contributor application deadline cited in the announcement was April 8, 2025 at 18:00 UTC, a historical date rather than a current deadline.
Can you test an SOF pipeline without physical DAI hardware?
The project’s intended answer is yes, for the portions supported by the Virtual DAI and its software stack. You can use a virtual endpoint, NoCodec and emulated-DMA-related integration to prototype stream behavior without attaching a physical DAI. The exact coverage depends on the platform, kernel, firmware, topology and driver versions you build. Physical hardware remains necessary for tests of real codec behavior, electrical interfaces, board clocks and other device-specific characteristics.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
