Responsibilities
Test and validate the performance and tuning of core voice processing blocks
Execute validation under diverse acoustic conditions, speaker configurations, and signal scenarios.
Collaborate with algorithm teams to provide data-driven feedback for tuning and optimization.
Conduct objective voice quality assessments using standardized metrics, including POLQA (ITU-T P.863) and PESQ (ITU-T P.862).
Organize and manage subjective "golden ear" listening panels to assess voice naturalness, intelligibility, and overall quality.
Define evaluation criteria and scoring methodologies aligned with industry standards and internal quality benchmarks.
Design and execute WER-based testing for scenarios involving voice transcription and speech recognition.
Correlate WER results with upstream audio processing quality to identify degradation sources.
Conduct testing in specialized laboratories that replicate real-world acoustic scenarios, including: Moving vehicles ,Noisy public environments – cafés, restaurants, crowded spaces Windy outdoor settings and Combined environmental stressors.
Measure and track Time to First Word (TTFW) and round-trip audio latency.
Establish performance baselines and monitoring frameworks for continuous tracking across firmware and software iterations.
Design and maintain automated test suites to detect audio artifacts – including glitches, pops, clicks, distortions, and dropouts.
Build regression frameworks that validate audio quality across new firmware and software updates.
Minimum Qualifications
BSc in Electrical Engineering, Computer Engineering, Acoustics, Audio Engineering, or a related field.
7+ years of experience in audio testing, voice quality evaluation, or acoustic system validation.
Proven experience working within or alongside professional audio/acoustic test laboratories.
Demonstrated ability to design repeatable, controlled test environments and methodologies.
Experience with both objective measurement tools and subjective listening evaluation protocols.
Strong system-level understanding with the ability to isolate quality issues to specific processing stages.
Deep understanding of audio signal processing fundamentals – frequency domain analysis, filtering, spectral characteristics, and time-domain behavior
Hands-on experience with voice quality measurement tools and standards (POLQA, PESQ, 3QUEST, ETSI standards).
Familiarity with HEAD acoustic test equipment – measurement microphones, artificial heads (HATS), mouth simulators, and anechoic/semi-anechoic chamber setups.
Strong proficiency in Python for test automation, data analysis, audio signal processing, and framework development.
Experience with audio acquisition systems, DACs/ADCs, and real-time audio streaming architectures.
Understanding of latency measurement methodologies in real-time audio systems.
Preferred Qualifications
MSc in Acoustics, Audio Signal Processing, Electrical Engineering, or a related discipline – considered a significant advantage.
Experience with speech recognition systems and WER evaluation methodologies.
Background in psychoacoustics or subjective audio evaluation research.
Experience with environmental simulation lab design and commissioning.
Knowledge of ITU-T P-series recommendations and ETSI test specifications.
