Conversation AI Audio Response in Sparkforge

By Sparkforge Support Team ·Last reviewed May 3, 2026

Audio Response lets Sparkforge Conversation AI accept voice notes and audio files on WhatsApp, Messenger, Instagram, and SMS/MMS. The bot transcribes the audio and replies based on your existing training.

Audio Response lets your Sparkforge Conversation AI bot hear customers, not just read them. When a contact sends a voice note or audio file, Sparkforge transcribes it, passes the text to your bot, and returns a context-aware reply. Customers talk naturally; you don't change anything about your training.

What it does

Conversation AI now accepts inbound audio across major messaging channels. Transcription happens behind the scenes, and the bot follows your existing settings — training, prompts, response mode, and timing — for consistent results.

Why it matters

Natural conversations — contacts speak instead of type.
Faster resolutions — automatic transcription feeds your trained bot for accurate replies.
Multi-audio intake — multiple files arrive as a single interaction.
Omnichannel reach — WhatsApp, Messenger, Instagram, and SMS/MMS share one workflow.
Consistent governance — audio replies respect your Wait Time and message-limit settings, just like text.

Supported audio types

Category	Supported	Notes
Voice notes (platform-native)	WhatsApp, Facebook, Instagram voice notes	Recorded with each app's mic; delivered as audio objects.
File formats (uploads)	OGG, MP3, MP4 audio-only, AAC, M4A, MPEG	Audio-only files. Video MP4s are not supported as audio inputs.
Multi-audio in one interaction	Yes	Multiple audio files sent in a short window are handled as a single interaction.

Channel compatibility

Facebook Messenger
Instagram Direct Messages
WhatsApp
SMS (MMS)

Each channel must already be connected and working with Conversation AI.

How to set up Audio Response

From your sub-account, go to AI Agents → Conversation AI → Agent List.
Click the three dots next to the bot you want to configure and pick Edit.
Toggle on Also allow this bot to respond to: Voice Notes.
Save your changes.
Send a voice note from a connected channel like WhatsApp to confirm the bot transcribes and replies correctly.

Behavior and limits

Wait Time aggregation — the bot waits the configured Wait Time Before Responding so it can collect multiple inbound messages (audio plus text) and send one unified reply.
Message limit — the Maximum Message Limit applies to audio just like text. Once reached, the bot sleeps until your standard reset.
Transcripts and transparency — review prompts, sources, and response info from the AI Response Info sidebar in Conversations.
Channel policies — Meta channels enforce policy windows (24-hour for Messenger and Instagram). Plan flows around them.

Frequently Asked Questions

Does Audio Response cost extra?

Usage falls under standard Conversation AI billing plus your channel's message fees (SMS/MMS, WhatsApp). Agencies can configure rebilling for Conversation AI usage.

Will the bot reply with audio or text?

Bots send standard channel messages, which are typically text. Design flows around text replies for maximum compatibility.

Can I limit audio handling to specific channels?

Yes. Assign only the channels you want the bot to use in Bot Settings. The bot will listen and respond only on assigned channels.

How are multiple audio files handled?

Multiple audio files sent within your Wait Time window get transcribed together and the bot crafts a single, context-aware reply.

Where do I review what the bot heard?

Open the AI Response Info sidebar in the conversation to see the transcribed prompt, response, and training sources used.