This document describes the process of setting up and running the Arm® Ethos™-U NPU Automatic Speech Recognition example.
Use-case code could be found in the following directory: source/use_case/asr.
The wav2letter automatic speech recognition model that is used with the code samples, expects audio data to be preprocessed in a specific way before performing an inference.
This section provides an overview of the feature extraction process used.
First, the audio data is normalized to the range (
Note: Mel-Frequency Cepstral Coefficients (MFCCs) are a common feature that is extracted from audio data and can be used as input for machine learning tasks such as keyword spotting and speech recognition. For implementation details, please refer to:
Next, a window of 512 audio samples is taken from the start of the audio clip. From these 512 samples, we calculate 13 MFCC features.
The whole window is shifted to the right by 160 audio samples and 13 new MFCC features are calculated. This process of shifting and calculating is repeated until enough audio samples to perform an inference have been processed.
In total, this is 296 windows that each have 13 MFCC features calculated for them.
After extracting MFCC features, the first and second order derivatives of these features, regarding time, are calculated.
These derivative features are then standardized and concatenated with the MFCC features (which also get standardized). At this point, the input tensor has a shape of 296x39.
These extracted features are quantized and an inference is performed.
For longer audio clips, where multiple inferences must be performed, then the initial starting position is offset by
(100*160) = 16000 audio samples. From this new starting point, MFCC and derivative features are calculated as before, until there is enough to perform another inference.
Padding can be used if there are not enough audio samples for at least one inference. This step is repeated until the whole audio clip has been processed. If there are not enough audio samples for a final complete inference, then the MFCC features are padded by repeating the last calculated feature until an inference can be performed.
Note: Parameters of the MFCC feature extraction all depend on what was used during model training. These values are specific to each model.
If you switch to a different ASR model than the one supplied, then the feature extraction process could be completely different to the one currently implemented.
The amount of time that audio samples that are offset for long audio clips is specific to the included wav2letter model.
After performing an inference, the raw output must be postprocessed to get a usable result.
The raw output from the model is a tensor of shape 148x29 where each row is a probability distribution over the possible 29 characters that can appear at each of the 148 time steps.
This wav2letter model is trained using context windows. This means that, depending on the bit of the audio clip that is currently being processed, only certain parts of the output are usable.
If this is the first inference, and multiple inferences are required, then ignore the final 49 rows of the output. Similarly, if this is the final inference from multiple inferences, then ignore the first 49 rows of the output.
Finally, if this inference is not the last, or the first inference, then ignore the first and last 49 rows of the model output.
Note: If the audio clip is small enough, then the whole of the model output is usable and there is no need to throw away any of the outputs before continuing.
Once any rows have been removed, the final processing can be done. To process the output, the letter with the highest probability at each time step is found first. Next, any letters that are repeated multiple times in a row are removed.
For example: [
p] becomes [
p]). Finally, the 29th blank token letter is removed from the output.
For the final output, the results from all inferences are combined before decoding. What you are left with is then displayed to the console.
In addition to the already specified build option in the main documentation, the Automatic Speech Recognition use-case adds:
asr_MODEL_TFLITE_PATH - The path to the NN model file in
TFLite format. The model is processed and then included into the application
axf file. The default value points to one of the delivered set of models. Note that the parameters
ETHOS_U_NPU_ENABLED must be aligned with the chosen model. In other words:
ETHOS_U_NPU_ENABLEDis set to
1, then the NN model is assumed to be optimized. The model naturally falls back to the Arm® Cortex®-M CPU if an unoptimized model is supplied.
ETHOS_U_NPU_ENABLEDis set to
0, then the NN model is assumed to be unoptimized. Supplying an optimized model in this case results in a runtime error.
asr_FILE_PATH: The path to the directory containing audio files, or a path to single WAV file, to be used in the application. The default value points to the
resources/asr/samples folder that contains the delivered set of audio clips.
asr_LABELS_TXT_FILE: The path to the text file for the label. The file is used to map letter class index to the text label. The default value points to the delivered
labels.txt file inside the delivery package.
asr_AUDIO_RATE: The input data sampling rate. Each audio file from
asr_FILE_PATH is preprocessed during the build to match the NN model input requirements. The default value is
asr_AUDIO_MONO: If set to
ON, then the audio data is converted to mono. The default value is
asr_AUDIO_OFFSET: Begins loading audio data and starts from this specified offset, defined in seconds. the default value is set to
asr_AUDIO_DURATION: The length of the audio data to be used in the application in seconds. The default is
0, meaning that the whole audio file is used.
asr_AUDIO_MIN_SAMPLES: Minimum number of samples required by the network model. If the audio clip is shorter than this number, then it is padded with zeros. The default value is
asr_MODEL_SCORE_THRESHOLD: Threshold value that must be applied to the inference results for a label to be deemed valid. The default is
asr_ACTIVATION_BUF_SZ: The intermediate, or activation, buffer size reserved for the NN model. By default, it is set to 2MiB and is enough for most models.
To ONLY build the automatic speech recognition example application, add
-DUSE_CASE_BUILD=asr to the
cmake command line, as specified in: Building.
Note: This section describes the process for configuring the build for the MPS3: SSE-300. To build for a different target platform, please refer to: Building.
To build only the automatic speech recognition example, create a build directory and navigate inside, like so:
mkdir build_asr && cd build_asr
On Linux, when providing only the mandatory arguments for CMake configuration, execute the following command to build only Automatic Speech Recognition application to run on the Ethos-U55 Fast Model:
cmake ../ -DUSE_CASE_BUILD=asr
To configure a build that can be debugged using Arm DS specify the build type as
Debug and then use the
Arm Compiler toolchain file:
cmake .. \ -DCMAKE_TOOLCHAIN_FILE=scripts/cmake/toolchains/bare-metal-armclang.cmake \ -DCMAKE_BUILD_TYPE=Debug \ -DUSE_CASE_BUILD=asr
For further information, please refer to:
Note: If re-building with changed parameters values, we recommend that you clean the build directory and re-run the CMake command.
If the CMake command succeeds, build the application as follows:
To see compilation and link details, add
Results of the build are placed under the
build/bin folder, like so:
bin ├── ethos-u-asr.axf ├── ethos-u-asr.htm ├── ethos-u-asr.map └── sectors ├── images.txt └── asr ├── ddr.bin └── itcm.bin
bin folder contains the following files:
ethos-u-asr.axf: The built application binary for the Automatic Speech Recognition use-case.
ethos-u-asr.map: Information from building the application. For example: The libraries used, what was optimized, and the location of objects.
ethos-u-asr.htm: Human readable file containing the call graph of application functions.
sectors/asr: Folder containing the built application. It is split into files for loading into different FPGA memory regions.
sectors/images.txt: Tells the FPGA which memory regions to use for loading the binaries in the
The application anomaly detection is set up to perform inferences on data found in the folder, or an individual file, that is pointed to by the parameter
To run the application with your own audio clips, first create a folder to hold them and then copy the custom clips into the following folder:
mkdir /tmp/custom_wavs cp my_clip.wav /tmp/custom_wavs/
Note: Clean the build directory before re-running the CMake command.
Next, when building, set
asr_FILE_PATH to the location of the following folder:
cmake .. \ -Dasr_FILE_PATH=/tmp/custom_wavs/ \ -DUSE_CASE_BUILD=asr
The audio flies found in the
asr_FILE_PATH folder are picked up and automatically converted to C++ files during the CMake configuration stage. They are then compiled into the application during the build phase for performing inference with.
The log from the configuration stage tells you what audio directory path has been used:
-- User option asr_FILE_PATH is set to /tmp/custom_wavs -- Generating audio files from /tmp/custom_wavs ++ Converting my_clip.wav to my_clip.cc ++ Generating build/generated/asr/include/InputFiles.hpp ++ Generating build/generated/asr/src/InputFiles.cc -- Defined build user options: -- asr_FILE_PATH=/tmp/custom_wavs
After compiling, your custom inputs have now replaced the default ones in the application.
Note: The CMake parameter
asr_AUDIO_MIN_SAMPLESdetermines the minimum number of input samples. When building the application, if the size of the audio clips is less than
asr_AUDIO_MIN_SAMPLES, then it is padded until it matches.
The application performs inference using the model pointed to by the CMake parameter
Note: If you want to run the model using an Ethos-U, ensure that your custom model has been successfully run through the Vela compiler before continuing.
For further information: Optimize model with Vela compiler.
To run the application with a custom model, you must provide a
labels_<model_name>.txt file of labels that are associated with the model. Each line of the file must correspond to one of the outputs in your model. Refer to the provided
labels_wav2letter.txt file for an example.
Then, you must set
asr_MODEL_TFLITE_PATH to the location of the Vela processed model file and
asr_LABELS_TXT_FILEto the location of the associated labels file.
cmake .. \ -Dasr_MODEL_TFLITE_PATH=<path/to/custom_model_after_vela.tflite> \ -Dasr_LABELS_TXT_FILE=<path/to/labels_custom_model.txt> \ -DUSE_CASE_BUILD=asr
Note: Clean the build directory before re-running the CMake command.
.tflite model file pointed to by
asr_MODEL_TFLITE_PATH, and the labels text file pointed to by
asr_LABELS_TXT_FILE are converted to C++ files during the CMake configuration stage. They are then compiled into the application for performing inference with.
The log from the configuration stage tells you what model path and labels file have been used, for example:
-- User option TARGET_PLATFORM is set to mps3 -- User option asr_MODEL_TFLITE_PATH is set to <path/to/custom_model_after_vela.tflite> ... -- User option asr_LABELS_TXT_FILE is set to <path/to/labels_custom_model.txt> ... -- Using <path/to/custom_model_after_vela.tflite> ++ Converting custom_model_after_vela.tflite to\ custom_model_after_vela.tflite.cc -- Generating labels file from <path/to/labels_custom_model.txt> -- writing to <path/to/build/generated/src/Labels.cc> ...
After compiling, your custom model has now replaced the default one in the application.
The FVP is available publicly from Arm Ecosystem FVP downloads.
For the Ethos-U evaluation, please download the MPS3 based version of the Arm® Corstone™-300 model that contains Cortex-M55 and offers a choice of the Ethos-U55 and Ethos-U65 processors.
To install the FVP:
Unpack the archive.
Run the install script in the extracted package:
Once the building has been completed, the application binary
ethos-u-asr.axf can be found in the
Assuming that the install location of the FVP was set to
~/FVP_install_location, then the simulation can be started by using:
A log output appears on the terminal:
telnetterminal0: Listening for serial connection on port 5000 telnetterminal1: Listening for serial connection on port 5001 telnetterminal2: Listening for serial connection on port 5002 telnetterminal5: Listening for serial connection on port 5003
This also launches a telnet window with the standard output of the sample application. It also includes error log entries containing information about the pre-built application version, TensorFlow Lite Micro library version used, and data types. The log also includes the input and output tensor sizes of the model compiled into the executable binary.
After the application has started, if
asr_FILE_PATH points to a single file, or even a folder that contains a single input file, then the inference starts immediately. If there are multiple inputs, it outputs a menu and then waits for input from the user.
User input required Enter option number from: 1. Classify next audio clip 2. Classify audio clip at chosen index 3. Run classification on all audio clips 4. Show NN model info 5. List audio clips Choice:
What the preceding choices do:
Classify next audio clip: Runs a single inference on the next in line.
Classify audio clip at chosen index: Runs inference on the chosen audio clip.
Note: Please make sure to select audio clip index within the range of supplied audio clips during application build. By default, a pre-built application has four files, with indexes from
Run ... on all: Triggers sequential inference executions on all built-in applications.
Show NN model info: Prints information about the model data type, input, and output, tensor sizes:
INFO - Model info: INFO - Model INPUT tensors: INFO - tensor type is INT8 INFO - tensor occupies 11544 bytes with dimensions INFO - 0: 1 INFO - 1: 296 INFO - 2: 39 INFO - Quant dimension: 0 INFO - Scale = 0.110316 INFO - ZeroPoint = -11 INFO - Model OUTPUT tensors: INFO - tensor type is INT8 INFO - tensor occupies 4292 bytes with dimensions INFO - 0: 1 INFO - 1: 1 INFO - 2: 148 INFO - 3: 29 INFO - Quant dimension: 0 INFO - Scale = 0.003906 INFO - ZeroPoint = -128 INFO - Activation buffer (a.k.a tensor arena) size used: 783168 INFO - Number of operators: 1 INFO - Operator 0: ethos-u
List audio clips: Prints a list of pair ... indexes. The original filenames are embedded in the application, like so:
[INFO] List of Files: [INFO] 0 => another_door.wav [INFO] 1 => another_engineer.wav [INFO] 2 => i_tell_you.wav [INFO] 3 => testing_routine.wav
Please select the first menu option to execute Automatic Speech Recognition.
The following example illustrates the output of an application:
INFO - Running inference on audio clip 0 => another_door.wav INFO - Inference 1/2 INFO - Inference 2/2 INFO - Final results: INFO - Total number of inferences: 2 INFO - For timestamp: 0.000000 (inference #: 0); label: and he walked immediately out of th INFO - For timestamp: 0.000000 (inference #: 1); label: e apartment by another door INFO - Complete recognition: and he walked immediately out of the apartment by another door INFO - Profile for Inference : INFO - NPU AXI0_RD_DATA_BEAT_RECEIVED beats: 6564262 INFO - NPU AXI0_WR_DATA_BEAT_WRITTEN beats: 928889 INFO - NPU AXI1_RD_DATA_BEAT_RECEIVED beats: 841712 INFO - NPU ACTIVE cycles: 28450696 INFO - NPU IDLE cycles: 476 INFO - NPU TOTAL cycles: 28451172
It can take several minutes to complete each inference. The average time is around 5-7 minutes, and on this audio clip, multiple inferences were required to cover the whole clip.
The profiling section of the log shows that for the first inference:
Ethos-U PMU report:
28,451,172 total cycle: The number of NPU cycles.
28,450,696 active cycles: The number of NPU cycles that were used for computation.
476 idle cycles: The number of cycles for which the NPU was idle.
6,564,262 AXI0 read beats: The number of AXI beats with read transactions from the AXI0 bus. AXI0 is the bus where
928,889 AXI0 write beats: The number of AXI beats with write transactions to the AXI0 bus.
841,712 AXI1 read beats: The number of AXI beats with read transactions from the AXI1 bus. AXI1 is the bus where the Ethos-U55 NPU reads the model. So, read-only.
For FPGA platforms, a CPU cycle count can also be enabled. However, do not use cycle counters for FVP, as the CPU model is not cycle-approximate or cycle-accurate.
The application prints the decoded output from each of the inference runs, and the final combined result.