Set voice starters to make prompt out of the transcript

.env.example

@@ -29,6 +29,7 @@ ALLOWED_TELEGRAM_USER_IDS=USER_ID_1,USER_ID_2
 # MAX_HISTORY_SIZE=15
 # MAX_CONVERSATION_AGE_MINUTES=180
 # VOICE_REPLY_WITH_TRANSCRIPT_ONLY=false
+# VOICE_REPLY_PROMPTS="Hi bot;Hey bot;Hi chat;Hey chat"
 # N_CHOICES=1
 # TEMPERATURE=1.0
 # PRESENCE_PENALTY=0.0

README.md

@@ -82,6 +82,7 @@ Check out the [Budget Manual](https://github.com/n3d1117/chatgpt-telegram-bot/di
 | `MAX_HISTORY_SIZE`                 | Max number of messages to keep in memory, after which the conversation will be summarised to avoid excessive token usage                                                                                                 | `15`                               |
 | `MAX_CONVERSATION_AGE_MINUTES`     | Maximum number of minutes a conversation should live since the last message, after which the conversation will be reset                                                                                                  | `180`                              |
 | `VOICE_REPLY_WITH_TRANSCRIPT_ONLY` | Whether to answer to voice messages with the transcript only or with a ChatGPT response of the transcript                                                                                                                | `false`                            |
+| `VOICE_REPLY_PROMPTS`              | A semicolon separated list of phrases ('Hi bot;Hello chat'). If the transcript starts with any of them - it will be treated as a prompt | -
 | `N_CHOICES`                        | Number of answers to generate for each input message. **Note**: setting this to a number higher than 1 will not work properly if `STREAM` is enabled                                                                     | `1`                                |
 | `TEMPERATURE`                      | Number between 0 and 2. Higher values will make the output more random                                                                                                                                                   | `1.0`                              |
 | `PRESENCE_PENALTY`                 | Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far                                                                                                         | `0.0`                              |

bot/main.py

@@ -68,6 +68,7 @@ def main():
         'stream': os.environ.get('STREAM', 'true').lower() == 'true',
         'proxy': os.environ.get('PROXY', None),
         'voice_reply_transcript': os.environ.get('VOICE_REPLY_WITH_TRANSCRIPT_ONLY', 'false').lower() == 'true',
+        'voice_reply_prompts': () if os.environ.get('VOICE_REPLY_PROMPTS', None) is None else os.environ.get('VOICE_REPLY_PROMPTS').split(";"),
         'ignore_group_transcriptions': os.environ.get('IGNORE_GROUP_TRANSCRIPTIONS', 'true').lower() == 'true',
         'group_trigger_keyword': os.environ.get('GROUP_TRIGGER_KEYWORD', ''),
         'token_price': float(os.environ.get('TOKEN_PRICE', 0.002)),

bot/telegram_bot.py

@@ -321,7 +321,10 @@ class ChatGPTTelegramBot:
                 if str(user_id) not in allowed_user_ids and 'guests' in self.usage:
                     self.usage["guests"].add_transcription_seconds(audio_track.duration_seconds, transcription_price)
 
-                if self.config['voice_reply_transcript']:
+                # check if transcript starts with any of the prefixes
+                response_to_transcription = any(transcript.startswith(prefix) for prefix in self.config['voice_reply_prompts'])
+
+                if self.config['voice_reply_transcript'] and not response_to_transcription:
 
                     # Split into chunks of 4096 characters (Telegram's message limit)
                     transcript_output = f"_{localized_text('transcript', bot_language)}:_\n\"{transcript}\""