Ensure you have Python installed on your machine, as it will be used for scripting the data transfer. Additionally, install the Elasticsearch Python client and DuckDB using pip:
```bash
pip install elasticsearch duckdb
```

Use the Python Elasticsearch client to connect to your Elasticsearch cluster. You’ll need the cluster's endpoint and any required authentication details:
```python
from elasticsearch import Elasticsearch

es = Elasticsearch(
hosts=["http://localhost:9200"], # Update with your Elasticsearch endpoint
http_auth=('user', 'password') # Add authentication if required
)
```

Query the Elasticsearch index to retrieve the data you want to transfer. Use the `search` API to fetch documents. Consider using scrolling if you're working with large datasets:
```python
index_name = 'your_index_name'
query = {
"query": {
"match_all": {}
}
}
response = es.search(index=index_name, body=query, scroll='2m', size=1000)

# Initialize variables for scroll loop
scroll_id = response['_scroll_id']
documents = response['hits']['hits']
```

Convert the raw documents from Elasticsearch into a format suitable for DuckDB, such as a list of dictionaries or a Pandas DataFrame:
```python
import pandas as pd

data = []
for doc in documents:
data.append(doc['_source'])

df = pd.DataFrame(data)
```

Initialize a connection to DuckDB and create a table schema that matches the data structure you retrieved from Elasticsearch. This can be done directly in Python:
```python
import duckdb

conn = duckdb.connect('my_database.duckdb')
create_table_query = """
CREATE TABLE IF NOT EXISTS my_table (
column1 TYPE,
column2 TYPE,
...
)
"""
conn.execute(create_table_query)
```

Use DuckDB's capabilities to insert the processed data. If you used a DataFrame, you can load it directly into DuckDB:
```python
conn.execute("INSERT INTO my_table SELECT FROM df")
```

Conduct a verification step to ensure that the data has been accurately transferred. Query the DuckDB table and compare it with the source data to validate:
```python
result = conn.execute("SELECT COUNT() FROM my_table").fetchall()
print("Number of records in DuckDB:", result[0][0])

# Compare with the number of documents retrieved from Elasticsearch
print("Number of documents retrieved:", len(data))
```

By following these steps, you can manually transfer data from Elasticsearch to DuckDB using Python without relying on external connectors or integrations. Adjust the scripts to cater to your specific data and environment requirements.