Google Dialogflow Benchmark

Google Dialogflow is a cloud-based natural language understanding (NLU) engine for building chatbots and voicebots. An agent can be trained by providing examples of what a user might say when interacting with it. It then analyzes and understands the user's intent based on the prior provided knowledge. Dialogflow uses Google speech to text technology behind the scenes to first transcribe speech and then analyze the extracted text to derive the user’s intent.

Prerequisites

• Ubuntu 20.04 (x86_64)
• Git
• Python
• PIP
• Google Cloud Platform Account

Usage

1. Clone the repository:

git clone https://github.com/Picovoice/speech-to-intent-benchmark.git

2. Install the dependencies:

pip3 install -r requirements.txt

3. Log in to Google cloud console, create a new project, and enable Dialogflow.

4. Create a new Dialogflow agent called barista

5. Use data/dialogflow/barista_432.zip to create and train the agent.

6. Run the benchmark:

python3 src/bench.py \
--engine GOOGLE_DIALOGFLOW \
--noise kitchen \
--google_dialogflow_credential_path ${YOUR_CREDENTIALS_PATH} \
--google_dialogflow_project_id ${YOUR_PROJECT_ID}

python3 src/bench.py \
--engine GOOGLE_DIALOGFLOW \
--noise kitchen \
--google_dialogflow_credential_path ${YOUR_CREDENTIALS_PATH} \
--google_dialogflow_project_id ${YOUR_PROJECT_ID}

Result