Use Ask FASS for live voice or text search and help appropriate to your account and workspace, with explicit microphone controls and an idle pause.
Quick answer
Open Ask FASS in the main search or snowflake launcher. Enable AI & Chatbot in privacy choices, then type or choose Talk to FASS. The assistant explains your allowance before microphone access or charging; if speech is unavailable, read it and choose Continue to voice.
Open Ask FASS from the main search area or the snowflake launcher, when available. Enable AI & Chatbot in privacy choices before sending a message. You can use ordinary search without AI.
The panel welcomes you and asks what you need help with. This fixed welcome does not make an AI request. Type your question and send it, or choose Talk to FASS. You can ask for schools, instructors, lessons, resorts or supported account help.
Before requesting your microphone or starting a voice connection, the assistant checks and explains your current free allowance or credit rates. This introduction uses no guest voice time or account credits. Where an English local device voice is available, it reads the explanation; that voice may differ from your selected AI conversation voice. If it cannot finish speaking, read the explanation and choose Continue to voice, or cancel and type instead. Changed credit rates require a new review, without an automatic connection retry.
The spoken introduction is brief. Choose Full voice introduction before connecting if you want to stop speech and review the complete explanation, then Continue to voice when ready. After a completed introduction, starting another call in the same open page uses a shorter reminder with your latest allowance and applicable prices. Refreshing the page or changing account, prices or trial status brings back the first-call explanation.
Allow microphone access when prompted. Wait for the short ready sound and Voice is ready message before speaking; the microphone remains off while connecting and during the sound. If your microphone is muted, unmute it when you want to speak. A browser without ready-sound support shows the ready message instead.
In voice, use Mute to pause your microphone, Stop reply to interrupt, and End voice to disconnect. You can keep typing after ending voice. Connected microphone audio is sent to OpenAI; “Listening” is not a promise that usage is free. Close the call when finished.
The panel can show that it is checking your request or current platform information. For a small set of supported questions, a labelled Draft reply appears as it is generated after the sources have been checked. Other replies appear when complete. Checked answers that are already ready appear immediately; the assistant does not add a typing delay.
Choose Stop reply to stop waiting. A draft is discarded if stopped or if the reply fails; it is not saved in the conversation or used to open pages. Links and actions become available only after the final reply finishes correctly. A draft does not mean a page has opened, a change has been saved or the problem is solved.
Deliberately stopping after draft text starts still uses one text request under your displayed allowance or credit rates. It does not add a second charge. See allowances and credits for failed-request handling.
After 90 seconds without meaningful conversation activity, the microphone pauses and asks whether to Continue voice or End voice. With no choice for another 30 seconds, voice disconnects and the current text history remains. Long replies and ongoing tasks are not treated as silence. A maximum call duration and remaining allowance/credit limit also apply. Reconnecting does not renew a guest trial or avoid a connection charge.
The assistant opens as a mobile panel. When it opens a page, compact controls leave the page visible; expand them to see the conversation. Minimising can keep voice connected. Locking the phone, switching apps, calls or network loss can interrupt browser audio; reconnect if shown as disconnected.
If microphone permission is denied, type instead. Use guided navigation or download the available transcript before resetting or leaving the page.
If audio cannot play, voice ends and offers Enable audio. This checks your device locally without starting another AI call. After it passes, choose Talk to FASS to start a new connection, or continue with text. If sound remains blocked, allow sound for this site in your browser settings. A brief network interruption may recover on the existing call; a disconnected call requires a deliberate new start with the displayed allowance and rates.
For unavailable voice, failed connections, missing replies or a support-reference message, try the checks in assistant troubleshooting before opening a ticket.
In Assistant settings, choose Hold to speak for manual microphone input. After voice is ready, hold the control, or hold Space/Enter while it is focused, and release to stop sending microphone input. Losing focus, muting, leaving the page, or cancelling releases the hold. Connected voice time still counts while released; choose End voice to stop the connection. This can help control input in noisy surroundings but cannot guarantee recognition or physically mute the device. Bounded spelling hints can use currently public favourite names from authorised preferences; this is not voice training.
Choose an offered voice before connecting. To change a voice once a call has started, end voice first. Review the remaining free trial or account-credit rates before starting; an on-screen countdown and low-time warning show the current allowance. At exhaustion use signup/sign-in or the displayed account options. You can still use normal search, download available text and contact support. See allowances and credits.
Still need help with this article?
Send the article URL and the step that needs more explanation.