Bengali Speech & Audio AI Specialist

Telus International Ai Inc; India; Remote; INR 600,000 - 900,000; Full time

Job summary

Telus International Ai Inc is seeking a Bengali Speech & Audio AI Specialist to evaluate and validate speech AI systems, ensuring natural-sounding voices, accurate ASR outputs, and high-quality datasets for large-scale user experiences.

You will contribute to TTS/ASR evaluation, audio editing, linguistic validation, dataset annotation, and model improvement across voice AI projects in a global setting. Native Bengali and professional English, plus hands-on audio tool experience, are required.

Qualifications

• Bachelor's degree in Audio Engineering, Sound Engineering, Music Technology, Linguistics, Phonetics, Speech-Language Pathology, Computational Linguistics, or CS with Speech/AI focus.
• OR 2+ years professional experience in audio production, dialogue editing, voice recording, podcast/audio post-production, speech dataset creation, or AI speech projects.
• Native-level Bengali (C1/C2+) – Required for pronunciation and linguistic accuracy.
• Professional English (B1+) – Required for technical instructions and documentation.
• Hands-on experience with at least one professional audio tool (e.g., Adobe Audition, iZotope RX, Pro Tools, Reaper, Audacity, Logic Pro, Cubase).
• Working knowledge of audio editing/restoration, noise reduction, spectrogram analysis, waveform interpretation, and audio quality assessment.

Responsibilities

• Evaluate AI-generated speech (TTS) for naturalness, pronunciation, rhythm, tone, and emotion.
• Identify translation errors, pronunciation issues, and linguistic nuances in ASR outputs.
• Conduct audio quality checks (loudness, noise, clipping, distortion, compliance).
• Analyze audio using waveform and spectrogram tools to identify quality issues.
• Provide structured feedback for model improvement.
• Perform professional audio editing (segmentation, trimming, noise reduction, enhancement).
• Apply audio restoration techniques to optimize speech samples.
• Prepare and validate speech content for training pipelines.
• Apply phonetics and linguistics to identify speech variations across accents and languages.
• Review voice recordings for clarity, consistency, and technical quality.
• Support creation and validation of high-quality speech datasets.
• Perform audio annotation and labeling for dataset integrity.
• Support TTS, ASR, voice similarity, voice generation, and conversational AI evaluation projects.
• Contribute to quality assurance workflows for voice technology projects.
• Document findings to inform model improvement.

Skills

Audio quality assessment

Linguistics knowledge

Quality assurance

Education

Bachelor's degree in Audio Engineering

2+ years relevant experience in audio/AI

Tools

Adobe Audition

iZotope RX

Pro Tools

Audacity

Logic Pro

Reaper

Cubase

Job description

About the Role

As a Bengali Speech & Audio AI Specialist in our Global AI Community, you evaluate and validate speech AI systems—ensuring machine-generated voices sound natural, speech recognition captures intent accurately, and voice AI models learn from high-quality datasets. Your work directly impacts millions of users voice-based AI experiences.

What You'll Do

Audio Evaluation & Quality Assessment

• Evaluate AI-generated speech (TTS) for naturalness, pronunciation, rhythm, tone, and emotion
• Identify tran ion errors, pronunciation issues, and linguistic nuances in speech recognition (ASR) outputs
• Conduct audio quality checks (loudness, noise, clipping, distortion, compliance)
• Analyze audio using waveform and spectrogram tools to identify quality issues
• Provide structured feedback for model improvement

Audio Editing & Technical Proficiency

• Perform professional audio editing (segmentation, trimming, noise reduction, enhancement)
• Apply audio restoration techniques to optimize speech samples
• Prepare and validate speech content for training pipelines

Speech Linguistics & Dataset Validation

• Apply phonetics and linguistics expertise to identify speech variations across accents and languages
• Review voice recordings for clarity, consistency, and technical quality
• Support creation and validation of high-quality speech datasets
• Perform audio annotation and labeling for dataset integrity

Voice AI Project Support

• Support TTS, ASR, voice similarity, voice generation, and conversational AI evaluation projects
• Contribute to quality assurance workflows for voice technology projects
• Document findings to inform model improvement

Qualification path

Mandatory Qualifications

Education / Professional Background

One of the following:

Bachelor's degree in Audio Engineering, Sound Engineering, Music Technology, Linguistics, Phonetics, Speech-Language Pathology, Computational Linguistics, or Computer Science (Speech/AI focus)

OR 2+ years professional experience in: audio production, dialogue editing, voice recording, podcast/audio post-production, speech dataset creation, or AI speech projects

Language Requirements

Native-level Bengali (C1/C2+) – Required for pronunciation and linguistic accuracy

Professional English (B1+) – Required for technical instructions and documentation

Technical Competency

Hands-on experience with at least one professional audio tool: Adobe Audition, iZotope RX, Pro Tools, Reaper, Audacity, Logic Pro, Cubase, or equivalent DAW

Working knowledge of: audio editing/restoration, noise reduction, spectrogram analysis, waveform interpretation, and audio quality assessment

Assessment & Verification Requirements

Completion of subject-specific qualification assessment (practical audio/speech evaluation)

Identity verification (IDV) process – Required as soon as possible after application completion