speech and Audio AI evaluation specialist

Innodata India; Dadri; Hybrid; INR 670,000 - 1,339,000; Full time

Job summary

Innodata India seeks an experienced Speech & Audio AI Evaluation Specialist to assess S2S, TTS, and real-time voice agents across global English accents. You will analyze intelligibility, cadence, pronunciation, and emotion, contributing to benchmarking and improvement of AI voice systems.

The role requires 2+ years in international voice evaluation within captive setups serving US/global clients, with strong phonetic or auditory background and multilingual capability.

Qualifications

• 2+ years of international voice experience exclusively within a Captive/In-House GCC serving US/global clients.
• Strong auditory ability to identify subtle vocal inflections, pitch, cadence, and articulation issues.
• Background in phonetics, phonology, auditory acoustics, or speech-language pathology.
• Experience in voice quality analytics, acoustic data annotation, or speech synthesis evaluation.
• Additional multilingual capability.

Responsibilities

• Evaluate S2S and TTS outputs for intelligibility, rhythm, and conversational cadence.
• Assess tone, empathy, hesitation, inection, irony, and emotional expression.
• Conduct pairwise audio preference evaluations.
• Verify lip-sync, multimodal synchronization, and acoustic scene consistency.
• Evaluate North American, British, Australian, and international English accents.
• Document audio anomalies, unnatural vocal artifacts, distortion, and phonetic errors.

Skills

Auditory skills

Speech evaluation

Multilingual capability

Education

Background in phonetics / phonology / auditory acoustics / speech-language pathology

Job description

Position: Speech & Audio AI Evaluation Specialist International Voice

Department: AI/LLM Operations Experience: 2+ Years Captive / GCC International Voice

Location: WFO/WFH

Employment: 3-Month Contract

Role Overview

Evaluate AI-powered Speech-to-Speech (S2S), Text-to-Speech (TTS), and real-time conversational voice agents. The role requires strong auditory skills to assess prosody, cadence, pronunciation, emotion, and other speech characteristics across global English dialects.

Key Responsibilities

• Evaluate S2S and TTS outputs for intelligibility, rhythm, and conversational cadence.
• Assess tone, empathy, hesitation, inection, irony, and emotional expression.
• Conduct pairwise audio preference evaluations.
• Verify lip-sync, multimodal synchronization, and acoustic scene consistency.
• Evaluate North American, British, Australian, and international English accents.
• Document audio anomalies, unnatural vocal artifacts, distortion, and phonetic errors.

Mandatory Requirements

• 2+ years of dedicated international voice experience exclusively within a Captive/In-House GCC serving US/global clients.
• Strong auditory ability to identify subtle vocal inections, pitch, cadence, and articulation issues. Preferred
• Background in phonetics, phonology, auditory acoustics, or speech-language pathology.
• Experience in voice quality analytics, acoustic data annotation, or speech synthesis evaluation.
• Additional multilingual capability.