v.0.1708.1 - small bugfixes in output parsing

Author: FlyingFathead

.dockerignore

@@ -0,0 +1 @@
+config/bot_token.txt

README.md

@@ -196,6 +196,13 @@ After launching your bot successfully, you can interact with it via Telegram (se
 - `/language` - set the model's transcription language (`auto` =  autodetect); if you know the language spoken in the audio, setting the transcription language manually with this command may improve both transcription speed and accuracy.
 
 ## Changes
+- v0.1708.1 - Small bug fixes in the output
+   - Note that running the program within `firejail` using Nvidia driver v.560.xx or newer requires i.e.:
+   ```
+   firejail --noblacklist=/sys/module --whitelist=/sys/module/nvidia* --read-only=/sys/module/nvidia* python src/main.py
+   ```
+   This is due to recent changes in Nvidia's driver handling on Linux, see i.e. [here](https://github.com/netblue30/firejail/issues/6509) or [here](https://github.com/netblue30/firejail/issues/6372)
+   - Dockerized versions should run without problems
 - v0.1708 - Direct video file uploads are now available
    - (to prevent abuse, they're disabled by default, see `config.ini`)
 - v0.1707 - New `config.ini` option: add sites that require full video download

src/main.py

@@ -3,7 +3,7 @@
 # openai-whisper transcriber-bot for Telegram
 
 # version of this program
-version_number = "0.1708"
+version_number = "0.1708.1"
 
 # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 # https://github.com/FlyingFathead/whisper-transcriber-telegram-bot/
@@ -311,11 +311,14 @@ async def process_queue(self):
                                                 ai_transcript_header = f"[ Transcript generated with: https://github.com/FlyingFathead/whisper-transcriber-telegram-bot/ | OpenAI Whisper model: `{model}` | Language: `{language}` ]"
                                                 header_content = f"{video_info_message}\n\n{ai_transcript_header}\n\n"
                                                 content = content[len(header_content):]
-                                            content = transcription_note + content  # Add transcription note
+                                            # content = transcription_note + content  # Add transcription note
 
                                         # Escape content before splitting
                                         content = html.escape(content)
 
+                                        # Then prepend the transcription note (do not escape it)
+                                        content = transcription_note + content  # Add transcription note
+
                                         # Define the maximum message length (Telegram limit is 4096)
                                         max_message_length = 3500