Voice input
Voice input allows respondents to dictate their answers using a microphone button. This feature is particularly useful for mobile users or situations where typing is inconvenient. ioZen processes the audio to ensure the final data is clean and correctly formatted.
Two processing modes
Section titled “Two processing modes”ioZen handles voice input differently depending on the type of field being filled.
Polish mode
Section titled “Polish mode”For text and textarea fields, ioZen applies a light polish to the dictated text. This process fixes capitalization, adds punctuation, and removes verbal fillers like “um” or “uh”. The goal is to produce a professional and readable response without changing the original meaning.
Each polish operation consumes 0.25 AI credits.
Extraction mode
Section titled “Extraction mode”For structured fields like email, phone, number, date, or select, ioZen uses a full extraction pipeline. The AI parses the spoken words into the specific data type required by the field.
For example, if a respondent says “next Friday at 3pm” in a date field, ioZen converts it into a standard ISO date. In a multi-select field, a response like “all except YouTube” correctly selects the appropriate options.
Each extraction operation consumes 1 AI credit.
Language and browser support
Section titled “Language and browser support”Voice input supports both English and Spanish. The system automatically detects the language based on your Intake Bot settings.
The microphone button only appears on browsers that support the Web Speech API. If a respondent uses a browser without microphone access, the button remains hidden to prevent confusion.
Voice input is available on all ioZen plans. Credits consumed count against your monthly allowance. For a full breakdown of allowances per plan, see AI credits.