Feat/ru random questions (#131)

* feat: random russian questions

* feat: dummy provides random russian questions

* fix: refactor questions

* fix: add pre-dummy phrase

* fix: add pre-dummy phrase

* fix: codestyle

* fix: path to file

response_selectors/convers_evaluation_based_selector/server.py

@@ -332,6 +332,7 @@ def rule_score_based_selection(dialog, candidates, scores, confidences, is_toxic
     best_id = np.argmax(curr_single_scores)
     best_candidate = candidates[best_id]
     best_skill_name = skill_names[int(best_id)]
+    prev_skill_names = [uttr["skill_name"] for uttr in dialog["bot_utterances"][-5:]]
 
     best_candidate = add_question_to_statement(
         best_candidate,
@@ -341,6 +342,7 @@ def rule_score_based_selection(dialog, candidates, scores, confidences, is_toxic
         link_to_question,
         link_to_human_attrs,
         not_sure_factoid,
+        prev_skill_names,
     )
 
     return best_candidate, best_id, curr_single_scores

response_selectors/convers_evaluation_based_selector/utils.py

@@ -21,6 +21,7 @@
 )
 
 sentry_sdk.init(getenv("SENTRY_DSN"))
+LANGUAGE = getenv("LANGUAGE", "EN")
 
 logging.basicConfig(format="%(asctime)s - %(name)s - %(levelname)s - %(message)s", level=logging.DEBUG)
 logger = logging.getLogger(__name__)
@@ -36,12 +37,20 @@
 misheard_with_spec2 = "like to chat about"
 alexa_abilities_spec = "If you want to use the requested feature say"
 
-LET_ME_ASK_YOU_PHRASES = [
-    "Let me ask you something.",
-    "I would like to ask you a question.",
-    "Hey, I have a quesiton to you.",
-    "May I ask you one interesting thing.",
-]
+LET_ME_ASK_YOU_PHRASES = {
+    "EN": [
+        "Let me ask you something.",
+        "I would like to ask you a question.",
+        "Hey, I have a quesiton to you.",
+        "May I ask you one interesting thing.",
+    ],
+    "RU": [
+        "Я бы хотела кое-что спросить.",
+        "О, у меня как раз есть вопрос для обсуждения.",
+        "Я хочу спросить тебя кое-о-чем интересном.",
+        "У меня есть кое-что интересное для обсуждения.",
+    ],
+}
 
 
 def join_used_links_in_attributes(main_attrs, add_attrs):
@@ -63,14 +72,15 @@ def add_question_to_statement(
     link_to_question,
     link_to_human_attrs,
     not_sure_factoid,
+    prev_skill_names,
 ):
 
     if not_sure_factoid and "factoid_qa" in best_skill_name:
         best_candidate["text"] = "I am not sure in my answer but I can try. " + best_candidate["text"]
     if best_candidate["text"].strip() in okay_statements:
         if dummy_question != "" and random.random() < ASK_DUMMY_QUESTION_PROB:
             logger.info(f"adding {dummy_question} to response.")
-            best_candidate["text"] += f"{np.random.choice(LET_ME_ASK_YOU_PHRASES)} {dummy_question}"
+            best_candidate["text"] += f"{np.random.choice(LET_ME_ASK_YOU_PHRASES[LANGUAGE])} {dummy_question}"
             # if this is not a link-to question, bot attributes will be still empty
             best_candidate["human_attributes"] = join_used_links_in_attributes(
                 best_candidate.get("human_attributes", {}), dummy_question_human_attr
@@ -82,6 +92,10 @@ def add_question_to_statement(
             best_candidate["human_attributes"] = join_used_links_in_attributes(
                 best_candidate.get("human_attributes", {}), link_to_human_attrs
             )
+    elif LANGUAGE == "RU" and best_skill_name == "dff_generative_skill":
+        if prev_skill_names[-3:] == 3 * ["dff_generative_skill"] and random.random() < ASK_DUMMY_QUESTION_PROB:
+            logger.info(f"adding russian {dummy_question} to dff-generative-skill response.")
+            best_candidate["text"] += f"{np.random.choice(LET_ME_ASK_YOU_PHRASES[LANGUAGE])} {dummy_question}"
 
     return best_candidate
 

skills/dummy_skill/README.md

@@ -0,0 +1,4 @@
+
+
+Russian Random questions are collected from https://mensby.com/women/relations/150-voprosov-chtoby-luchshe-uznat-sobesednika-ili-sobesednicu
+and https://habr.com/ru/company/testutor/blog/298180/

skills/dummy_skill/connector.py

@@ -77,6 +77,9 @@
 with open("skills/dummy_skill/nounphrases_facts_map.json", "r") as f:
     NP_FACTS = json.load(f)
 
+with open("skills/dummy_skill/russian_random_questions.txt", "r") as f:
+    RUSSIAN_RANDOM_QUESTIONS = f.readlines()
+
 
 class RandomTopicResponder:
     def __init__(self, filename, topic, text):
@@ -271,6 +274,11 @@ async def send(self, payload: Dict, callback: Callable):
                 attrs += [{"type": "link_to_for_response_selector"}]
                 human_attrs += [human_attr]
                 bot_attrs += [{}]
+            elif is_russian:
+                cands += [random.choice(RUSSIAN_RANDOM_QUESTIONS)]
+                attrs += [{"type": "link_to_for_response_selector"}]
+                human_attrs += [{}]
+                bot_attrs += [{}]
 
             if not is_russian:
                 facts_same_nps = []