Voice activity detection

Phonexia voice-activity-detection microservice specializes in identifying segments within recordings that capture voice or speech. To learn more, visit the technology's home page.

Installation

Docker image

Getting the image

You can easily obtain the voice activity detection image from docker hub. There are 2 variants of the image. For CPU and for GPU.

CPU

You can get the CPU image by specifying a direct version in the tag (e.g. 1.0.0) or latest for the latest image:

docker pull phonexia/voice-activity-detection:latest

GPU

The GPU images has a suffix -gpu in the image tag (e.g. 1.0.0-gpu) or you can use a tag gpu to get the latest version. In these images, the most computationally demanding tasks are handled by the GPU. The prerequisites are NVIDIA GPU with drivers and nvidia-container-toolkit installed (see the Installing the NVIDIA Container Toolkit for more info).

docker pull phonexia/voice-activity-detection:gpu

Running the image

You can start the microservice and list all the supported options by running:

docker run --rm -it phonexia/voice-activity-detection:latest --help

The output should look like this:

voice-activity-detection [OPTIONS]


OPTIONS:
  -h,     --help              Print this help message and exit 
  -m,     --model file REQUIRED (Env:PHX_MODEL_PATH) 
                              Path to a model file. 
  -k,     --license_key string REQUIRED (Env:PHX_LICENSE_KEY) 
                              License key. 
  -a,     --listening_address address [[::]]  (Env:PHX_LISTENING_ADDRESS) 
                              Address on which the server will be listening. Address '[::]' 
                              also accepts IPv4 connections. 
  -p,     --port number [8080]  (Env:PHX_PORT) 
                              Port on which the server will be listening. 
  -l,     --log_level level:{error,warning,info,debug,trace} [info]  (Env:PHX_LOG_LEVEL) 
                              Logging level. Possible values: error, warning, info, debug, 
                              trace. 
          --keepalive_time_s number:[0, max_int] [60]  (Env:PHX_KEEPALIVE_TIME_S) 
                              Time between 2 consecutive keep-alive messages, that are sent if 
                              there is no activity from the client. If set to 0, the default 
                              gRPC configuration (2hr) will be set (note, that this may get the 
                              microservice into unresponsive state). 
          --keepalive_timeout_s number:[1, max int] [20]  (Env:PHX_KEEPALIVE_TIMEOUT_S) 
                              Time to wait for keep alive acknowledgement until the connection 
                              is dropped by the server. 
          --device TEXT:{cpu,cuda} [cpu]  (Env:PHX_DEVICE) 
                              Compute device used for inference. 
          --num_threads_per_instance NUM (Env:PHX_NUM_THREADS_PER_INSTANCE) 
                              Number of threads per instance (applies to CPU processing only). 
                              Use N CPU threads in the microservice for each request. Number of 
                              threads is automatically detected if set to 0. 
          --num_instances_per_device NUM:UINT > 0 (Env:PHX_NUM_INSTANCES_PER_DEVICE) 
                              Number of instances per device (both CPU and GPU processing). 
                              Microservice can process requests concurrently if value is >1. 
          --device_index ID (Env:PHX_DEVICE_INDEX) 
                              Device identifier 
          --profile TEXT:{speech_to_text,voice_biometrics} [speech_to_text]  (Env:PHX_PROFILE) 
                              Specify the profile to use for Voice Activity Detection. Choose 
                              'speech_to_text' to use configuration for speech-to-text 
                              scenarios or 'voice_biometrics' to use configuration for voice 
                              activity detection used in voice biometric technologies. 

note

The model and license_key options are required. To obtain the model and license, contact Phonexia.

You can specify the options either via command line arguments or via environmental variables.

Notice the --profile option. This option allows you to select a configuration profile used by other Phonexia technologies. Voice activity detection than produces the same speech segmentation as in these technologies. The default value, speech_to_text, corresponds to the profile optimized for the Speech-to-Text technology, while the voice_biometrics profile is tailored for biometric technologies, such as Speaker Identification.

CPU

Run the container with the mandatory parameters:

docker run --rm -it -v /opt/phx/models:/models -p 8080:8080 /phonexia/voice-activity-detection:latest --model /models/generic-3.2.0.model --license_key ${license-key}

GPU

To run GPU images you will need to make the GPU available inside the docker container. This is done by --gpus parameter (typically --gpus all), see the Access an NVIDIA GPU chapter for more info, for example:

Run the container with the mandatory parameters:

docker run --rm -it --gpus all -v /opt/phx/models:/models -p 8080:8080 /phonexia/voice-activity-detection:gpu --model /models/generic-3.2.0.model --license_key ${license-key}

Replace the /opt/phx/models, generic-3.2.0.model and license-key with the corresponding values.

With this command, the container will start, and the microservice will be listening on port 8080 on localhost.

Docker compose

Docker compose

There are 2 variants of the docker image. For CPU and for GPU. Create a docker-compose.yml file for the specific variant:

CPU

version: '3'
services:
  voice-activity-detection:
    image: phonexia/voice-activity-detection:latest
    environment:
      - PHX_MODEL_PATH=/models/generic-3.2.0.model
      - PHX_LICENSE_KEY=<license-key>
    ports:
      - 8080:8080
    volumes:
      - ./models:/models/

GPU

version: '3'
services:
  voice-activity-detection:
    image: phonexia/voice-activity-detection:gpu
    environment:
      - PHX_MODEL_PATH=/models/generic-3.2.0.model
      - PHX_LICENSE_KEY=<license-key>
    ports:
      - 8080:8080
    volumes:
      - ./models:/models/
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]

Create a models folder in the same directory as the docker-compose.yml file and place a model file in it. Replace <license-key> with your license key and generic-3.2.0.model with the actual name of a model.

note

The model and license_key options are required. To obtain the model and license, contact Phonexia.

You can than start the microservice by running:

$ docker compose up

Helm chart

The optimal way for large scale deployment is by using container orchestration system. Take a look at out Helm chart deployment page for deployment using Kubernetes.

Performance optimization

The voice-activity-detection microservice supports GPU acceleration.

In the docker images with GPU support, the GPU acceleration is enabled by default. While GPU acceleration will be used primarily, certain processing tasks will still rely on CPU resources.

For better performance, multiple microservices can share a GPU unit. The number of microservice instances per GPU depends on the hardware used.

Microservice communication

gRPC API

For communication, our microservices use gRPC, which is a high-performance, open-source Remote Procedure Call (RPC) framework that enables efficient communication between distributed systems using a variety of programming languages. We use an interface definition language to specify a common interface and contracts between components. This is primarily achieved by specifying methods with parameters and return types.

Take a look at our gRPC API documentation. The voice-activity-detection microservice defines a VoiceActivityDetection service with remote procedure called Detect. The Detect procedure accepts an argument (also referred to as "message") called DetectRequest, which contains the audio as an array of bytes.

This DetectRequest argument is streamed, meaning that it may be received in multiple requests, each containing a part of the audio. Once all requests have been received and processed, the Detect procedure returns a message called DetectResponse which consists of the resulting segmentation.

Connecting to microservice

There are multiple ways how you can communicate with our microservices.

Generated library

Using generated library

The most common way how to communicate with the microservices is via a programming language using a generated library.

Python library

If you use Python as your programming language, you can use our official gRPC Python library.

To install the package using pip, run:

pip install phonexia-grpc

You can then import:

• Specific libraries for each microservice that provide the message wrappers.
• stubs for the gRPC clients.

from phonexia.grpc.common.core_pb2 import Audio, RawAudioConfig, TimeRange
from phonexia.grpc.technologies.voice_activity_detection.v1.voice_activity_detection_pb2 import DetectRequest, DetectResponse
from phonexia.grpc.technologies.voice_activity_detection.v1.voice_activity_detection_pb2_grpc import VoiceActivityDetectionStub

Generate library for programming language of your choice

For the definition of microservice interfaces, we use the standard way of protocol buffers. The services, together with the procedures and messages that they expose, are defined in the so-called proto files.

The proto files can be used to generate client libraries in many programming languages. Take a look at protobuf tutorials for how to get started with generating the library in the languages of your choice using the protoc tool.

You can find the proto files developed by Phonexia in this repository.

Using existing clients

Python client

Phonexia Python client

The easiest way to get started with testing is to use our simple Python client. To get it, run:

pip install phonexia-voice-activity-detection-client

After the successful installation, run the following command to see the client options:

voice_activity_detection_client --help

grpcurl client

grpcurl client

If you need a simple tool for testing the microservice on the command line, you can use grpcurl. This tool can serialize and send a request for you, if you provide the request body in JSON format and specify the endpoint.

The audio content in the body must be encoded in Base64. The request also cannot exceed 4 MiB, therefore it's necessary to split bigger files to multiple chunks. You can use jq tool to generate JSON input for grpcurl.

Now you can make the request. The microservice supports reflection. That means that you don't need to know the API in advance to make a request. Replace ${path_to_audio_file} with corresponding value.

base64 -w 4000000 ${path_to_audio_file} | jq -cnR '{"audio":{"content":inputs}}' | grpcurl -plaintext -use-reflection -d @ localhost:8080 phonexia.grpc.technologies.voice_activity_detection.v1.VoiceActivityDetection/Detect

The grpcurl automatically serializes the segments in the response to this request into a JSON format.

GUI clients

GUI clients

If you'd prefer to use a GUI client like Postman or Warthog to test the microservice, take a look at the GUI Client page in our documentation. Note that you will still need to convert the audio into the Base64 format manually as those tools do not support it by default either.

Further links

• Maintained by Phonexia
• Contact us via e-mail, or open a ticket at the Phonexia Service Desk
• File an issue
• See list of licenses
• See the terms of use

Versioning

We use Semantic Versioning.