Algorithmic-Human-Development-Group/Multilingual-Therapy-Dialogues

![arXiv]() ![GitHub]()

Dataset Summary

Multilingual Therapy Dialogues is a diverse and bilingual dataset consisting of paired dialogues between patients and therapists in both Persian and English.

Dataset Statistics

• —Number of samples: 7,179

English:

1. 1.Average tokens per sentence: 101.30
2. 2.Maximum tokens in a sentence: 939
3. 3.Average characters per sentence: 567.85
4. 4.Number of unique tokens: 32,968

Persian:

1. 1.Average tokens per sentence: 100.06
2. 2.Maximum tokens in a sentence: 1,413
3. 3.Average characters per sentence: 516.57
4. 4.Number of unique tokens: 33,298

Dataset Fields

1. 1.Patient: Original English text spoken by the patient.
2. 2.Therapist: Original English text spoken by the therapist.
3. 3.Translated Patient: Persian translation of the patient's text.
4. 4.Translated Therapist: Persian translation of the therapist's text.

Dataset Generation Pipeline

The dataset was constructed using the following steps:

1. 1.Data Collection: Dialogues were collected from various public sources, including:
2. 2.Mental Health Counseling Conversations
3. 3.Mental Health CSV Dataset
4. 4.Mental Health Conversational Data
5. 5.Additional manually curated sources

1. 1.Translation: English dialogues were translated into Persian using the SeamlessM4T model by Meta AI.

1. 1.Refinement: Translations were revised and enhanced in three steps using GPT-4o:
2. 2.First pass to make the tone more natural and emotionally sympathetic to be more likely to real world scenarios.
3. 3.Second pass to improve fluency and human-likeness.
4. 4.Final pass for consistency and correction of subtle translation errors.

1. 1.Filtering: Only meaningful and conte

Usage Instructions

Option 1: Manual Download

Visit the dataset repository and download the SAT_dataset.csv file.

Option 2: Programmatic Download

Use the huggingface_hub library to download the dataset programmatically:

from huggingface_hub import hf_hub_download
import pandas as pd

dataset = hf_hub_download(
    repo_id="Algorithmic-Human-Development-Group/Multilingual-Therapy-Dialogues",
    filename="SAT_dataset.csv",
    repo_type="dataset"
)
df = pd.read_csv(dataset)
df.head()

Citations

If you find our paper, code, data, or models useful, please cite the paper:

To be updated once the paper is published.

Contact

If you have questions, please email sinaaelahimanesh@gmail.com or mahdi.abootorabi2@gmail.com.