diff --git a/.idea/mypy.xml b/.idea/mypy.xml
new file mode 100644
index 0000000000000000000000000000000000000000..7492791335b2316fdb290a77882fb322cba1b406
--- /dev/null
+++ b/.idea/mypy.xml
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+ <component name="MypyConfigService">
+ <option name="customMypyPath" value="$USER_HOME$/miniconda3/envs/asr-benchmarks/bin/mypy" />
+ </component>
+</project>
\ No newline at end of file
diff --git a/sziszapangma/core/transformer/web_embedding_transformer.py b/sziszapangma/core/transformer/web_embedding_transformer.py
new file mode 100644
index 0000000000000000000000000000000000000000..af56681b334c8ab92a79100d5502576330855179
--- /dev/null
+++ b/sziszapangma/core/transformer/web_embedding_transformer.py
@@ -0,0 +1,38 @@
+import json
+from typing import Dict, List, Optional
+
+import numpy as np
+import requests
+
+from sziszapangma.core.transformer.embedding_transformer import EmbeddingTransformer
+
+
+class WebEmbeddingTransformer(EmbeddingTransformer):
+ _lang_id: str
+ _host: str
+ _auth_token: Optional[str]
+
+ def __init__(self, lang_id: str, host: str, auth_token: Optional[str]):
+ self._lang_id = lang_id
+ self._host = host
+ self._auth_token = auth_token
+
+ def get_embedding(self, word: str) -> np.ndarray:
+ return self._make_request([word])[word]
+
+ def get_embeddings(self, words: List[str]) -> Dict[str, np.ndarray]:
+ return self._make_request(words)
+
+ def _make_request(self, words: List[str]) -> Dict[str, np.ndarray]:
+ print(f"call for words: {json.dumps(words)}")
+ headers = {"Content-Type": "application/json"}
+ if self._auth_token is not None:
+ headers["Authorization"] = f"Bearer {self._auth_token}"
+ result = requests.post(
+ f"{self._host}/embeddings/{self._lang_id}", headers=headers, json=words
+ )
+ if not result.ok:
+ print(f"response_code {result.status_code}")
+ result_text = result.text
+ parsed_result = json.loads(result_text)
+ return {it: np.array(parsed_result[it]) for it in parsed_result.keys()}
diff --git a/sziszapangma/integration/service_core/__init__.py b/sziszapangma/integration/service_core/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/sziszapangma/integration/service_core/asr/__init__.py b/sziszapangma/integration/service_core/asr/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/sziszapangma/integration/service_core/asr/asr_base_processor.py b/sziszapangma/integration/service_core/asr/asr_base_processor.py
new file mode 100644
index 0000000000000000000000000000000000000000..f0763f957df6dd8826b0b6028f80ec376e0b8448
--- /dev/null
+++ b/sziszapangma/integration/service_core/asr/asr_base_processor.py
@@ -0,0 +1,67 @@
+import os
+import traceback
+import uuid
+from abc import ABC, abstractmethod
+from pathlib import Path
+from typing import Optional
+
+from flask import Flask, Response, jsonify, request
+from flask_httpauth import HTTPTokenAuth
+
+from sziszapangma.integration.service_core.asr.asr_result import AsrResult
+
+_TEMP_DIRECTORY = "asr_processing"
+_AUTH_TOKEN = "AUTH_TOKEN"
+_SERVICE_PORT = "SERVICE_PORT"
+
+
+class AsrBaseProcessor(ABC):
+ user_token: str
+
+ def __init__(self):
+ self.user_token = os.environ[_AUTH_TOKEN]
+
+ @abstractmethod
+ def process_asr(self, audio_file_path: str) -> AsrResult:
+ """Method to call for ASR results."""
+
+ def process_request(self) -> Response:
+ file_tag = str(uuid.uuid4())
+ f = request.files["file"]
+ if f is not None and f.filename is not None:
+ file_extension = f.filename.split(".")[-1]
+ file_name = f"{file_tag}.{file_extension}"
+ file_path = f"{_TEMP_DIRECTORY}/{file_name}"
+ f.save(file_path)
+ try:
+ transcription = self.process_asr(file_path)
+ os.remove(file_path)
+ result_object = jsonify(
+ {"transcription": transcription.words, "full_text": transcription.full_text}
+ )
+ except Exception as exception:
+ print(exception)
+ traceback.print_exc()
+ result_object = jsonify({"error": "Error on asr processing"})
+ else:
+ result_object = jsonify({"error": "Error on asr processing"})
+ return result_object
+
+ def is_token_correct(self, token: str) -> Optional[str]:
+ if token == self.user_token:
+ return "asr_client"
+ else:
+ return None
+
+ def health_check(self) -> Response:
+ return jsonify({"status": "running"})
+
+ def start_processor(self):
+ app = Flask(__name__)
+ auth = HTTPTokenAuth(scheme="Bearer")
+ auth.verify_token(self.is_token_correct)
+ Path(_TEMP_DIRECTORY).mkdir(parents=True, exist_ok=True)
+ app.route("/process_asr", methods=["POST"])(auth.login_required(self.process_request))
+ app.route("/health_check", methods=["GET"])(self.health_check)
+ port = int(os.environ[_SERVICE_PORT]) if _SERVICE_PORT in os.environ else 5000
+ app.run(debug=True, host="0.0.0.0", port=port)
diff --git a/sziszapangma/integration/service_core/asr/asr_result.py b/sziszapangma/integration/service_core/asr/asr_result.py
new file mode 100644
index 0000000000000000000000000000000000000000..cdf7c208f29f4585116191b67769a4d26150c3da
--- /dev/null
+++ b/sziszapangma/integration/service_core/asr/asr_result.py
@@ -0,0 +1,8 @@
+from dataclasses import dataclass
+from typing import List
+
+
+@dataclass(frozen=True)
+class AsrResult:
+ words: List[str]
+ full_text: str
diff --git a/sziszapangma/integration/service_core/embedding/__init__.py b/sziszapangma/integration/service_core/embedding/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/sziszapangma/integration/service_core/embedding/embedding_base_processor.py b/sziszapangma/integration/service_core/embedding/embedding_base_processor.py
new file mode 100644
index 0000000000000000000000000000000000000000..60e1250d8fc24059f9be9a2730deeb787ef7babd
--- /dev/null
+++ b/sziszapangma/integration/service_core/embedding/embedding_base_processor.py
@@ -0,0 +1,55 @@
+import os
+from abc import ABC, abstractmethod
+from typing import List, Optional
+
+import numpy as np
+import numpy.typing as npt
+from flask import Flask, Response, jsonify, request
+from flask_httpauth import HTTPTokenAuth
+
+_TEMP_DIRECTORY = "asr_processing"
+_AUTH_TOKEN = "AUTH_TOKEN"
+_SERVICE_PORT = "SERVICE_PORT"
+
+
+class EmbeddingBaseProcessor(ABC):
+ user_token: str
+
+ def __init__(self):
+ self.user_token = os.environ[_AUTH_TOKEN]
+
+ @abstractmethod
+ def get_embedding(self, phrase: str, language: str) -> npt.NDArray[np.float64]:
+ """Method to call for embedding of phrase."""
+
+ @staticmethod
+ def numpy_to_list(arr) -> List[float]:
+ return [float(it) for it in arr]
+
+ def process_embeddings(self, language: str):
+ words = request.json
+ if words is not List[str]:
+ raise Exception("Incorrect body")
+ return jsonify(
+ {word: self.numpy_to_list(self.get_embedding(word, language)) for word in words}
+ )
+
+ def is_token_correct(self, token: str) -> Optional[str]:
+ if token == self.user_token:
+ return "asr_client"
+ else:
+ return None
+
+ def health_check(self) -> Response:
+ return jsonify({"status": "running"})
+
+ def start_processor(self):
+ app = Flask(__name__)
+ auth = HTTPTokenAuth(scheme="Bearer")
+ auth.verify_token(self.is_token_correct)
+ app.route("/process_embedding", methods=["POST"])(
+ auth.login_required(self.process_embeddings)
+ )
+ app.route("/health_check", methods=["GET"])(self.health_check)
+ port = int(os.environ[_SERVICE_PORT]) if _SERVICE_PORT in os.environ else 5000
+ app.run(debug=True, host="0.0.0.0", port=port)