Speech Software Engineer

About this role

ASAPP seeks a Senior Speech Software Engineer to optimize and scale real-time voice AI infrastructure for enterprise call centers. You'll bridge cutting-edge speech research with production systems, tuning ASR/TTS models for low-latency, high-accuracy performance while architecting resilient, high-concurrency voice pipelines.

What you'll do

• Tune and optimize ASR and TTS models for accuracy, noise robustness, and real-world call center environments
• Architect scalable, high-availability voice infrastructure handling thousands of concurrent real-time audio streams
• Design and operate end-to-end streaming pipelines integrating ASR, LLM, and TTS for live conversations
• Define speech quality evaluation frameworks and build monitoring dashboards for production performance
• Evaluate and integrate emerging speech technologies like noise suppression, VAD, and diarization
• Collaborate with Speech Scientists, ML Researchers, and Product teams to productionize models and translate requirements

What they're looking for

• Golang or Python
• ASR and TTS systems (applied or production experience)
• Low-latency, high-concurrency distributed systems design
• Speech quality metrics (WER, CER, MOS, latency)
• Audio fundamentals (codecs, buffering, packet loss, jitter)
• ML model evaluation and fine-tuning
• Real-time media or streaming data handling
• Kubernetes, Docker, and cloud deployment

Benefits

• Work on cutting-edge AI-powered voice technology at scale
• Collaborate with world-class Speech Scientists and ML Researchers
• Fast-paced startup environment with continuous learning opportunities
• Global team with hubs in NYC, Mountain View, Latin America, and India
• Performance-based bonus compensation