Skip to main content
LLM ReadyRaw Markdown for AI agents:contacts-and-audio-files.mdFull index:llms.txtllms-full.txt

Contacts & Audio Files

Centralize customer caller identity across your workspace with Contacts & Address Books, and manage studio-recorded greetings, announcements, and queue hold music in Audio Files.


Contacts & Address Books​

Whenever an inbound call or SMS arrives—or an agent searches the directory before placing an outbound call—Firetell normalizes the phone number to E.164 format and matches it against your workspace Contacts in real time.

1. Organizing Address Books & Access Scope​

Every contact belongs to an Address Book. Click Address Books in the top-right corner of Console → Call Center → Contacts to create and structure directories by visibility scope:

Address Book ScopeWho Can Access ItTypical Use Cases
Workspace-WideAll agents, supervisors, Call Flows, and Voice AI Agents in the workspace.Master customer directory, vendor hotlines, and shared VIP lookup lists.
Team DirectoryMembers assigned to a specific Agent Team (e.g., Enterprise Sales or Billing).Department-specific leads, regional client portfolios, or partner escalations.
Personal Agent BookPrivate to an individual agent account.Dedicated account manager contact lists and personal follow-up callbacks.

Mark your primary directory as the Default Address Book so newly created contacts (via Agent Portal or API) land in the right repository automatically.

2. Contact Profile Fields & Multi-Number Normalization​

Click New Contact to create a rich customer record containing:

  • Identity & Organization: First Name, Last Name, Company, and Job Title.
  • Multiple Phone Numbers: Add mobile, work, home, fax, or other numbers and designate a Primary number. Firetell indexes normalized E.164 digits automatically so +1 (415) 555-0199 and 14155550199 resolve to the same customer profile.
  • Multiple Emails: Store work, personal, or other email addresses.
  • Segmentation & Context: Attach searchable Tags (e.g., vip, enterprise, renewal-q4), free-form agent Notes, customer Timezone, preferred Language, Assigned Agent, and custom key-value CRM metadata.

3. How Contacts Power Live Calls & Voice AI​

  • Agent Workspace Screen-Pop: Displays the caller's full name, company, tags, and notes before the agent answers an inbound ring.
  • Voice AI Personalization: Automatically populates the {{caller_name}} runtime variable in Autonomous Voice AI Agents so AI receptionists greet returning customers by name.
  • Unified Call & SMS History: Links historical Call Logs (CDRs) and Two-Way SMS threads to the customer's record.
  • CRM Synchronization: Sync records two-way via the Contacts REST API, Address Books REST API, and real-time contact.created / contact.updated / contact.deleted webhooks.

Audio Files & IVR Prompts​

Audio assets uploaded under Console → Call Center → Audio Files provide custom voice greetings, holiday/after-hours announcements, ringback tones, and queue hold music across your Visual Call Flow Builder (IVR).

1. Uploading & Managing Studio Prompts​

  1. Navigate to Console → Call Center → Audio Files and click Upload Audio.
  2. Enter a clear, searchable Title using a consistent naming convention—for example:
    • IVR_Main_Welcome_EN
    • IVR_AfterHours_Closed_VI
    • Queue_HoldMusic_AcousticLoop
  3. Upload your audio file. Firetell automatically processes telephony-ready sample rates (8 kHz G.711 for PSTN/SIP carrier compatibility alongside wideband streams) and displays file duration and storage size in the table.
  4. Click the inline Play preview button on any row to verify audio clarity and volume before binding the file to a live Call Flow node.

Phone networks compress dynamic range much more aggressively than desktop speakers. Follow these studio targets so your IVR prompts sound crisp on mobile phones, landlines, and WebRTC softphones alike:

Audio PropertyRecommended TargetWhy It Matters
Container / FormatUncompressed .wav (16-bit PCM) or high-bitrate .mp3 (192+ kbps)Uploading lossless .wav prevents generation loss when transcoded to telephony codecs (PCMU/PCMA/Opus).
Sample Rate16 kHz or 44.1/48 kHz source (auto-resampled to 8 kHz for PSTN legs)Preserves wideband clarity on WebRTC/SIP calls while downsampling cleanly for traditional PSTN carriers.
Channel LayoutMono (1 channel)All telephony call legs are mono. Downmix stereo tracks to mono beforehand to avoid phase cancellation when left/right channels sum together.
Integrated Loudness-16 LUFS (Voice) / -20 LUFS (Hold Music) with -1.5 dBTP true peakPrevents harsh clipping on mobile earpieces and avoids jarring volume jumps when transitioning from hold music to a live agent's voice.
Background Music Under SpeechAvoid—or duck at least -18 dB below voiceoverTelephony codecs (G.711) prioritize speech frequencies (300 Hz–3.4 kHz); loud background music muddies spoken menu options (Press 1, Press 2).
Updating Live Call Flow Prompts

Before deleting an audio file in Console → Call Center → Audio Files, verify that no active Play Audio or Queue Hold Music node in your published Call Flows still references that asset. For automated campaign announcements, see the Audios REST API.