Using Artificial Intelligence

Copilot AI

Copilot AI is FSFO’s First Officer AI experience. It supports natural flight-deck conversation, Copilot requests, and spoken AI replies.

The First Officer profile is separate from the Cabin Crew profile. You can use different voices, languages, accents, speech pace, and custom prompts for the Copilot and Flight Attendants.

Important: OpenAI is required for AI understanding and conversational responses. Microsoft Azure Speech may generate the spoken voice, but it cannot power the AI conversation itself.

AI Access Modes

Mode	What it does
Standard FSFO	Uses local regional voice packs only. This mode does not provide Live AI conversation or AI-generated content.
My API Key	Uses your own OpenAI API key for AI understanding and responses. You may choose OpenAI Voice or Microsoft Azure Speech for spoken output.
FSFO Credits	Uses FSFO-provided access and your FSFO credit balance. A personal API key is not required.

Get an OpenAI API Key

You need an OpenAI API key only when you select My API Key. You do not need a personal key when using FSFO Credits.

1. Go to OpenAI API Keys.
2. Sign in to your OpenAI Platform account, or create one.
3. Select Create new secret key.
4. Give the key a recognizable name, such as FSFO AI & Voices.
5. Copy the key immediately and store it securely.
6. In FSFO, open Settings > AI & Voices.
7. Select My API Key.
8. Paste the key into the OpenAI API Key field.
9. Select Test Connection, then select Save AI Settings.

Important: Your full OpenAI secret key is displayed only when it is created. Never share it in screenshots, Discord posts, support requests, or public documents. If you lose it, create a new key and update FSFO.

Note: A ChatGPT subscription and OpenAI API billing are separate services. API usage is charged through your OpenAI Platform account.

AI Access Mode options in Settings > AI & Voices.

Configure Copilot AI

1. Open Settings > AI & Voices.
2. Under AI Access Mode, select My API Key or FSFO Credits.
3. If using your own API key, enter and test your OpenAI API key.
4. Under AI & Voice Engine, choose OpenAI Voice or Microsoft Azure Speech.
5. If using Azure Speech, enter the Azure Speech key and region or endpoint.
6. Under Voice Pack Role, select First Officer.
7. Choose the First Officer Language, Region / Accent, Voice, and Speech Pace.
8. Choose an AI Style (Personality).
9. Adjust Creativity and Response Length if desired.
10. Add a Custom First Officer Prompt if you want the Copilot to follow specific instructions or communication preferences.
11. Enable Copilot conversation if you want to ask questions and have a natural conversation using the assigned Live AI button.
12. Select Save AI Settings.

Copilot Voice Settings

Setting	Purpose
Language	Selects the language used by the selected First Officer voice.
Region / Accent	Selects the available regional or accent option for the selected provider and language.
Voice	Selects the voice used for spoken Copilot replies.
Speech Pace	Adjusts the playback speed of the selected voice.
AI Style (Personality)	Sets the overall tone of AI-generated responses, such as Professional Airline, Warm and Friendly, or Premium International.
Creativity	Controls how predictable or varied AI-generated wording is.
Response Length	Sets the preferred maximum length of generated responses.
Custom First Officer Prompt	Adds role-specific instructions for Copilot behavior and phrasing.

Copilot Conversation

Enable Copilot conversation to use the Live AI button for natural questions and discussion.

Examples include:

• Asking for a flight-status summary.
• Requesting operational information.
• Asking the Copilot to explain a checklist or procedure.
• Having a natural spoken conversation with the First Officer AI.

OpenAI handles the AI understanding and response generation. Azure Speech, when selected, can provide the spoken voice but does not replace OpenAI for the AI conversation.

Copilot Voice Pack Builder

The Voice Pack Builder creates only the missing WAV files for the selected role and voice.

To create a Copilot voice pack:

1. Select First Officer under Voice Pack Role.
2. Select the desired provider, language, accent, and voice.
3. Select Generate Missing First Officer WAVs.

FSFO stores the generated voice pack in the Sounds folder. Existing clips are preserved, and only missing clips are generated.

AI Settings Reset

Reset to Defaults on the AI & Voices page resets only AI-related settings, including AI access mode, provider selection, role voice settings, style, prompts, conversation settings, and voice-pack preferences.

It does not reset General Settings, Audio & Speech device settings, Cabin Crew Announcement settings, or other user preferences.

Use Save AI Settings after making changes.

Cabin Crew

Cabin Crew settings control passenger announcements, cabin-to-flight-deck intercom replies, Cabin Crew AI content, and generated Flight Attendant voice packs.

The Cabin Crew profile is independent from the Copilot / First Officer profile. Changing the Cabin Crew voice does not change the Copilot voice.

Important: FSFO controls when an announcement or intercom reply is used. The selected playback source controls how it sounds.

Configure the Cabin Crew Voice

1. Open Settings > AI & Voices.
2. Under Voice Pack Role, select Cabin Crew.
3. Choose the Cabin Crew Language, Region / Accent, Voice, and Speech Pace.
4. Choose an AI Style (Personality) if you use AI-generated Cabin Crew content.
5. Adjust Creativity and Response Length if desired.
6. Add a Custom Cabin Crew Prompt if you want the Flight Attendant AI to use a specific tone or style.
7. Select Preview Voice to hear the selected Cabin Crew voice.
8. Select Save AI Settings.

Cabin Crew Voice Settings

Setting	Purpose
Language	Selects the language used by the selected Cabin Crew voice.
Region / Accent	Selects the available regional or accent option for the selected provider and language.
Voice	Selects the voice used for Cabin Crew AI announcements, replies, and generated voice packs.
Speech Pace	Adjusts the playback speed of the Cabin Crew voice.
Custom Cabin Crew Prompt	Adds role-specific instructions for the Flight Attendant AI, such as tone, wording, and service style.

Cabin Crew Announcements

The Cabin Crew Announcements area controls passenger-facing announcements and cabin intercom replies.

Each announcement can be:

• Enabled or disabled.
• Previewed using its configured playback source.
• Assigned an automatic trigger and delay.
• Edited with static audio, TTS text, or an AI prompt.
• Used as a built-in announcement or as a custom announcement.

Cabin Crew announcement and response settings.

Cabin Crew Playback Sources

Source	What it does
Static	Plays WAV files. Built-in announcements use their built-in audio. Supported Cabin Crew replies can use a matching generated Cabin Crew voice pack when enabled.
TTS	Speaks configured text using the selected Cabin Crew Windows SAPI voice.
AI	Generates dynamic announcements or replies using OpenAI, the Cabin Crew voice profile, the AI style, and the configured prompt.

Cabin Crew Intercom Replies

Cabin Comms Reply Source controls how cabin crew replies are voiced over the flight-deck intercom.

Reply Source	Behavior
Static	Uses matching generated Cabin Crew WAVs when enabled and available. If no matching generated clip is available, FSFO falls back to the Cabin Crew regional static voice.
TTS	Uses the selected Cabin Crew Windows SAPI voice.
AI	Generates a dynamic Cabin Crew reply using the Cabin Crew AI profile.

When Static is selected, enable When Static is selected, use matching generated Cabin Crew WAVs if you want FSFO to first look in the generated Cabin Crew voice pack.

This is useful when you want static intercom replies to sound like the selected Flight Attendant AI voice.

Cabin Crew Voice Pack Builder

The Voice Pack Builder can create a dedicated Flight Attendant WAV pack.

1. Select Cabin Crew under Voice Pack Role.
2. Select the desired provider, language, accent, and voice.
3. Select Generate Missing Cabin Crew WAVs.

FSFO generates only missing clips and preserves any clips already present in the selected Cabin Crew voice pack.

The generated WAVs are stored in the FSFO Sounds folder and may be used by supported static Cabin Crew replies.

Announcement List

The Announcement List contains built-in announcement rows and one optional custom announcement row.

The top controls are:

• Enable All — turns the Cabin Crew announcement system on or off without deleting individual row settings.
• Responses — switches the grid between passenger announcements and intercom response definitions.
• Announcements — returns from Responses view to the announcement list.
• + Add Announcement — opens the custom announcement editor.

The announcement grid columns are:

• # — display order.
• Announcement — announcement name.
• Trigger — selected automatic trigger.
• Delay — delay before playback.
• Segment — flight segment used for trigger logic.
• Conditions — operational note or phase reminder.
• Status / Actions — enable checkbox plus Preview, Trigger, and Edit buttons.

The row actions are:

• Row checkbox — enables or disables that announcement.
• Preview — plays a preview using the selected source and current voice settings.
• Trigger — opens the trigger editor.
• Edit — opens the appropriate editor for the selected playback source.

Built-In Announcement Rows

#	Announcement	Default Segment	Default Trigger	Default Delay	Purpose
1	Boarding Welcome	Preflight	Boarding Started	00:45	Welcomes passengers once boarding begins.
2	Boarding Complete Call	Preflight	Boarding Complete	00:05	Announces that boarding is complete.
3	Doors Armed	Preflight	Doors Closed	00:05	Doors armed and cross-check complete.
4	Departure Safety Briefing	Preflight	Doors Closed	00:40	Departure safety briefing.
5	Service Start	Airborne	Service Started	00:00	Announces the start of cabin service.
6	Top of Descent	Descent	Start of Descent	00:00	Begins arrival preparation.
7	Cabin Secure for Landing	Approach	Below 10,000 ft	00:20	Final cabin-secure reminder before landing.
8	Arrival Welcome	Taxi-In	Runway Exit	00:20	Welcomes passengers to the destination.
9	Doors Disarmed	Shutdown	Engines Off	00:05	Announces doors disarmed and cross-check complete.
10	Custom Announcement	All	Boarding Started	00:10	User-defined announcement that appears after a custom announcement name is saved.

Trigger Editor

Select Trigger on an announcement row to open the trigger editor.

The trigger editor contains:

• Trigger Point — the event FSFO watches for.
• Delay — the delay before playback.
• Selected Trigger — readback of the selected trigger.

If an announcement is manual-only, the trigger selector is disabled and shows Manual Only.

Segment	Available triggers
Preflight	Boarding Started, Boarding Complete, Doors Closed, Pushback Start, Beacon On
Taxi-Out	Pushback Complete, Beacon On
Climb	Above 10,000 ft, Above 18,000 ft, Top of Climb
Airborne	Above 18,000 ft, Top of Climb, Start of Descent, TOD
Descent	Start of Descent, -1000, -2000 to -5000, Below 10,000 ft
Approach / Final	Below 15,000 ft through Below 5,000 ft, Landing Lights On
Rollout / Taxi-In	Runway Exit, Landing Lights Off, Ground Speed Below 45 MPH, Ground Speed Below 20 MPH
Shutdown	Doors Open, Engines Off, Beacon Off, Taxi Off

Special trigger notes:

• Service Start supports Service Started.
• Cabin Secure for Landing supports the approach trigger set.
• Arrival Welcome supports Runway Exit, Landing Lights Off, Ground Speed Below 45 MPH, and Ground Speed Below 20 MPH.
• Doors Disarmed supports Beacon Off, Taxi Off, Engines Off, and Doors Open.
• An enabled announcement using Boarding Complete plays only when FSFO detects boarding completion.
• If the custom announcement remains on the All segment, it behaves as Manual Only until a supported segment is selected.

Responses

Select Responses to switch from passenger announcements to intercom response definitions.

Responses are used when:

• The flight deck calls the cabin.
• The cabin calls the flight deck.
• A communications-panel button starts an intercom exchange.
• A supported voice-command workflow requests a reply.

Common Cabin Crew response types include:

• Boarding Complete / Still In Progress
• Start Boarding acknowledged
• Ready for Takeoff / Not Ready
• Release Seats acknowledged
• Cabin Service starting
• Turbulence Ahead secured
• Delay Advisory acknowledged
• Turbulence Cleared / Resuming Service
• Prepare for Landing acknowledged
• Prepare for Takeoff acknowledged
• Ready for Landing / Not Ready
• Coffee Request acknowledged
• Cockpit Door Cover acknowledged

Common Flight Deck response types include:

• Boarding Request approved or standby
• Boarding Complete acknowledged
• Deboarding Request approved or standby
• Ready for Takeoff Call acknowledged
• Ready for Landing Call acknowledged
• Cabin Service Request approved or standby

Response-specific placeholders:

• {status_reason} — reason for a not-ready or delayed response.
• {response_reason} — alternate reason placeholder accepted by the response system.

Custom Announcements

Select + Add Announcement to create one custom announcement.

The custom editor fields are:

• Announcement Name — required.
• Trigger — automatic trigger for the selected segment.
• Delay — playback delay, such as 00:10.
• Segment — determines which automatic triggers are available.
• Conditions — free-form operational note shown in the grid.
• Save Announcement — saves the custom announcement and displays it in the list.
• Cancel — closes the editor without saving.

Important custom-announcement behavior:

• One custom announcement is currently supported.
• The selected segment controls whether automatic triggers are available.
• The custom announcement stores its own name, segment, trigger, delay, conditions, text, AI prompt, enabled state, and optional static WAV path.
• Use the Edit button after saving to configure its text, AI prompt, or custom WAV file.

Content Editing

The editor changes according to the configured playback source.

AI

• Edit the AI prompt sent to OpenAI.
• The Cabin Crew AI profile, style, creativity, response length, and custom Cabin Crew prompt affect the generated result.
• AI requires OpenAI access through your own API key or FSFO Credits.

TTS

• Edit the exact text that FSFO will speak.
• FSFO uses the selected Cabin Crew Windows SAPI voice.

Static

• Built-in announcements and responses use their assigned WAV files.
• The custom announcement can use a custom WAV file through the Browse WAV... option.
• Supported static Cabin Crew replies can use matching generated Cabin Crew WAVs when that option is enabled.

Common placeholders supported in TTS messages and AI prompts:

• {airline}
• {flight_number}
• {origin}
• {destination}
• {dep_city}
• {arr_city}
• {aircraft}
• {cruise_altitude}
• {weather}
• {time}
• {local_time}

Best Practices

• Use Standard FSFO and Static playback for a local, zero-AI setup.
• Use TTS when you want exact wording with a Windows SAPI voice.
• Use AI when you want dynamic Cabin Crew announcements and responses.
• Configure the First Officer and Cabin Crew roles separately so the Copilot and Flight Attendants have distinct voice identities.
• Generate a Cabin Crew voice pack and enable generated static Cabin Crew WAVs if you want static intercom replies to match the Flight Attendant AI voice.
• Use Boarding Complete for announcements that should play only after boarding has fully completed.
• For later arrival timing, use Landing Lights Off, Ground Speed Below 45 MPH, or Ground Speed Below 20 MPH instead of Runway Exit.
• For a reliable post-arrival door-disarm message, Engines Off is usually the safest default.
• Keep API keys private and do not share them in screenshots, support requests, or public posts.