Skip to content

AI-CallMaker v0.1

Latest
Latest

Choose a tag to compare

View all tags
@knightfoxapps knightfoxapps released this 28 Mar 20:53
· 4 commits to main since this release
Application
5c6c1c4
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Absolutely! Here's a clean and professional set of Release Notes for your project, tailored for version v1.0.0.

📦 Release Notes — v1.0.0

Release Date: March 27, 2025
Project: AI Call Center Conversation Generator & Audio Synthesizer

✨ New Features

  • OpenAI GPT-4 Integration
    Automatically generates realistic 20+ exchange customer service conversations using natural language.

  • Amazon Polly Voice Synthesis
    Converts conversations into stereo audio with lifelike neural voices.

  • Stereo Channel Assignment
    Agent voice on left channel, customer on right, for authentic call simulation.

  • Voice Gender Matching
    Automatically assigns male/female Polly voices based on agent and customer names.

  • Account Number Handling
    Numbers are spaced for proper digit-by-digit pronunciation in audio output.

  • Custom Setup Wizard
    Prompts for API keys, company info, products, customer locations, and more — with ability to skip or randomize topics.

  • Escalation Scenarios & Moods
    Generates conversations with varying tones: polite, angry, frustrated — and handles escalations.

  • Robust Error Handling
    Retry logic for Polly, debug mode with printouts, audio integrity checks, and export confirmation.

🛠 Improvements

  • Handles Polly’s 3000-character limit by chunking long segments cleanly.
  • More informative debug output for audio byte lengths and API metadata.
  • Directory structure created automatically (call_audios/).
  • User prompts improved for clarity and flexibility.

🐛 Known Issues

  • Some extremely short or overly complex exchanges may be truncated in Polly.
  • Mood matching is randomized — no real sentiment control in voice tone (yet).
  • Occasionally, audio may cut short if Polly API silently fails despite 200 response (still under monitoring).

🔜 Coming Soon

  • Batch JSON export of call metadata and transcripts.
  • UI frontend for non-technical users.
  • Multi-lingual support for international call centers.
  • Speaker diarization and subtitle generation.

Let me know if you want this in a changelog format (CHANGELOG.md) too, or styled for GitHub Releases.

Assets 2
Loading