How to optimize the transcription of your audio and video know-how (Transcription settings file)

This section of the documentation describes the Transcription settings.gdoc file that allows you to optimize the process of transcribing audio and video files into text form for your BuddyPro instance. By setting these parameters correctly, you can significantly improve the quality and accuracy of the transcripts while also influencing the processing cost of your Know-How.

You can find the Transcription settings.gdoc file in the root (main) folder of your BuddyPro instance on Google Drive (i.e., the same level where the sources and ROLES folders are located). This is a simple text document (ideally a Google Doc) where each line defines one parameter and its value in the Parameter: value format.

9.1 The “Improve transcriptions” parameter

• What it does: This parameter determines whether BuddyPro performs additional "enhancements" to the text using more advanced AI analysis after the basic verbatim transcription of the audio or video file.
• Possible values:
  • true (usually the default setting):
    • After standard transcription, BuddyPro will further process the text. The result is better structured and more readable text: filler words (various "uh", "actually", "as if") are removed, minor typos are corrected, sentences can be shortened or better rephrased, and sometimes headings or bullets are added for better clarity.
    • This "improved" transcription generally provides better input for further training of your instance's AI (e.g. for role generation or for more accurate knowledge base searches during conversations).
    • ⚠️ Note: Enabling this feature(Improve transcriptions: true) significantly increases AI credit consumption when processing your audio/video files with the /update command. This is because it is an additional, computationally intensive AI operation.
    • You can check the total training cost to date (processing and transcription of all your sources) by using the new /stats command on the AI costs of Buddy Pro training line. This will help you better evaluate the impact of the Improve transcriptions setting on your total costs.
  • false:
    • BuddyPro will only perform a standard, as faithful as possible, verbatim transcription of the audio/video recording to text. This text will not be further automatically edited or enhanced.
    • This option is significantly cheaper in terms of AI credit consumption.
    • It is suitable if:
      • You have a very large amount of audio/video material and minimizing the cost of initial processing is a priority.
      • Your recordings are already of very high quality, with clear and intelligible speech without a lot of exaggeration or filler words.
      • You prefer to have a completely verbatim transcript available for your own manual review and editing.
• How to change:
  1. Open the Transcription settings.gdoc file on your Google Drive.
  2. Find the line starting with Improve transcriptions:.
  3. Change the value after the colon to true or false (without the quotes). For example: Improve transcriptions: false.
  4. Save the changes in the file.
  5. The change will be reflected the next time the audio/video files are processed with /update (i.e., for newly added files or files that have been changed since the last time they were processed).
• Consider the costs:

	Audio/video files	YouTube/Instagram videos	Other types of content
Improve transcriptions: true	~0.023$ / minute	~0.03$ / minute	N/A
Improve transcriptions: false	~0.01$ / minute	~0.017$ / minute	~0.003$ / standard page (1800 characters)

9.2 LANGUAGEparameter

• What it does: Helps the AI more accurately identify the language of your audio and video recordings, which directly contributes to the higher quality and accuracy of the resulting transcription.
• Possible values:
  • Blank (usually the default setting):
    • If you leave this line blank (e.g. Language: without any value after the colon), BuddyPro will attempt to automatically detect the language for each individual audio/video file.
    • This option works well if you have material recorded in different languages or you are not 100% sure of the language of some recordings.
  • Two-letter ISO 639-1 language code:
    • If all your audio/video materials (or the current batch of files you are recording and want to have transcribed) are in one particular language, you can specify its two-letter ISO 639-1 code here. This directly instructs the AI what language to expect when transcribing, which can make the result more accurate.
    • Example codes: en English, de German, es Spanish, it Italian. A complete list of codes can be found, for example, on Wikipedia.
• How to change:
  1. Open the file Transcription settings.gdoc.
  2. Find the Language: line.
  3. After the colon, type the appropriate ISO 639-1 language code (e.g. Language: en) or leave the field blank for automatic detection.
  4. Save the file.
• Use when recording material in different languages:
  • It is easiest to leave the Language: field blank; BuddyPro will attempt to detect the language.
  • Alternatively, if you want maximum accuracy, you can upload materials in language batches:
    1. Set Language: es (for example).
    2. Upload all Spanish audio/video files to the sources/Files folder.
    3. Run /update.
    4. When finished, change the setting to Language: en.
    5. Upload all English audio/video files.
    6. Run /update again.

9.3 Parameter KEYWORDS

• What it does: Allows you to define a list of specific words, multi-word phrases, proper nouns, product names, or unique terms that you want to increase the likelihood of their correct and accurate transcription from audio/video recordings. The AI will focus more on these keywords during the transcription process.
• Why it's useful: Many technical terms, project or product names, person names, or specific company or community designations are not part of the common vocabulary that large language models have been trained with. Artificial intelligence could therefore misspell them, not recognize them correctly, or transcribe them phonetically incorrectly.
• How to enter:
  1. Open the Transcription settings.gdoc file.
  2. Find the Keywords: line.
  3. After the colon, type your keywords or multi-word phrases, separated by a comma. Individual keywords or phrases (if multi-word) should be in the exact form you want them transcribed (including upper/lower case if the written form matters).
     • Keywords: Pavel Riha,BuddyPro,FutureBot,Online Strategist,Wealth Dynamics,Sales Automators
     • In this example, "Pavel Riha", "BuddyPro", "FutureBot", "Online Strategist", "Wealth Dynamics" and "Sales Automators" are the individual key items. Some are single-word, others multi-word - spaces within the multi-word phrase are fine. A comma separates each key item.
  4. Save the file.
• Recommendation:
  • Add here the names of key people in your business, your community, or authors you frequently reference.
  • Include the names of your main products, services, online courses, methodologies, or frameworks.
  • Include any other unique terms, abbreviations, or specific designations that are important to you to capture accurately and consistently in your transcripts.

**Important Note:**Changes to the Transcription settings.gdoc file are automatically applied to newly processed audio/video files the next time you run the /update command. Existing transcriptions are not automatically updated by these changes.

Voice Message Recommendation

BuddyPro can also transcribe voice messages, but it’s important to keep them within the limit.

👉 Voice messages should be a maximum of 15 minutes long. Longer recordings will not be transcribed and will not be considered in responses.

For longer content, we recommend using a different format – for example, upload the audio/video file to the sources/Files folder on Google Drive.

---