The Fifth Elephant
The Fifth Elephant

The Fifth Elephant 2024 Annual Conference (12th &13th July)

Maximising the Potential of Data — Discussions around data science, machine learning & AI

All submissions

Rishav Kumar

@a1rishav

Samvaadini - a telecalling voice bot for blue-collar hiring in India

Submitted May 6, 2024

Session type: Demo - showcase of your work

In this talk, we introduce ‘Samvaadini’, a telecalling voice bot designed specifically for hiring blue-collar workers in India. Samvaadini leverages state-of-the-art AI to capture human responses adeptly, autonomously determine the next action, and respond in a voice indistinguishable from a human.

Audience for this talk
The primary audience for this talk includes AI researchers, developers, and engineers, as well as HR professionals and business leaders interested in innovative recruitment technologies.

The problem you are trying to solve with Samvaadini
Samvaadini aims to optimize the recruitment process for blue-collar jobs, particularly the difficulty of conducting effective telephonic interviews in noisy environments and at scale.

Scope of the talk

  1. Demo showcasing Samvaadini’s capabilities.
  2. Discussion of the unique challenges faced when building and deploying a voice bot for blue-collar recruitment, including handling noisy environments and varied dialects.
  3. An exploration of the business and efficiency gains achieved through the deployment of Samvaadini.

How will participants benefit from your talk?

  1. A comprehensive understanding of the complexities involved in creating a robust voice bot for telecalling in challenging environments.
  2. Knowledge of practical AI applications in HR and recruitment.
  3. Insights into the potential business impacts and efficiency improvements driven by advanced AI technologies.
  4. Inspiration and ideas for developing similar AI-driven solutions in their respective fields.

Format of the session : PPT

All submissions

Comments

Login to leave a comment

  • Gagan Gupta

    @gagangupta Editor

    Congratulations, Rishav. Your submission has been accepted.

    The next step is to update the following:

    1. Architectural Overview: Detailed discussion on the architecture of Samvaadini. What technologies and frameworks were used in building the voice bot?
    2. Speech-to-Text (STT) and Text-to-Speech (TTS) Models: Insights into how the STT models were trained, built, and orchestrated. Similarly, how was the TTS system developed and managed?
    3. Handling Multi-Lingual Support: If there was any multi-lingual support, please elaborate on how it was implemented.
    4. Technical Challenges: A comprehensive discussion on the unique technical challenges faced (e.g., handling noisy environments, varied dialects) and the specific solutions employed to tackle these issues.
    Posted 9 months ago
    • RK

      Rishav Kumar

      @a1rishav Submitter

      @gagangupta
      Thank you for your queries. Please find my responses below:

      1. Architectural Overview: Unfortunately, I cannot disclose details about the architecture due to a confidentiality clause.
      2. Technologies and Frameworks: Similarly, I am unable to provide information on the specific technologies and frameworks used in building Samvaadini due to confidentiality restrictions.
      3. Handling Multi-Lingual Support: Samvaadini understands English and Hindi but responds in Hindi only.
      4. Technical Challenges:
        Background Noise: Affects STT accuracy.
        Language-Specific Nuances: Requires careful tuning of LLMs.
        Accent Issues: Extensive training needed for various dialects.
      Posted 9 months ago (edited 9 months ago)

Hosted by

The Fifth Elephant
The Fifth Elephant
Jump starting better data engineering and AI futures

Supported by

Gold Sponsor

Atlassian
Atlassian
Atlassian unleashes the potential of every team. Our agile & DevOps, IT service management and work management software helps teams organize, discuss, and compl

Silver Sponsor

Google
Google
Together, we can build for everyone.

Workshop sponsor

Datastax
Datastax
Datastax, the real-time AI Company.

Lanyard Sponsor

Uber
Uber
We reimagine the way the world moves for the better.

Sponsor

Monster API
Monster API
MonsterAPI is an easy and cost-effective GenAI computing platform designed for developers to quickly fine-tune, evaluate and deploy LLMs for businesses.

Community Partner

FOSS United Foundation
FOSS United Foundation
FOSS United is a non-profit foundation that aims at promoting and strengthening the Free and Open Source Software (FOSS) ecosystem in India. more

Beverage Partner

BONOMI
BONOMI
BONOMI is a ready to drink beverage brand based out of Bangalore. Our first segment into the beverage category is ready to drink cold brew coffee.