Appen

• Review audio files and transcribe spoken content accurately • Refine transcripts to ensure high-quality data • Transcribing audio recordings into text with high accuracy • Capturing speech exactly as spoken, including pauses, filler words, repetitions, and incomplete sentences • Adding timestamps to match the audio • Identifying and labeling different speakers in conversations • Marking non-speech events such as laughter, background noise, or unclear audio

• Write prompts in your native language designed to test an AI model. • Evaluate the AI's responses using a structured rating guide. • Identify areas where the AI's responses can be improved. • Translate your prompts and evaluations into English.

Role Description Help improve speech AI by validating Vietnamese audio recordings. You'll listen to audio samples and select the correct labels based on provided guidelines, helping build better AI-powered speech recognition for your native language. - Evaluate Vietnamese audio recordings by selecting the provided labels - Apply validation guidelines precisely and consistently - Assess accent and pronunciation accuracy specific to the language variant Qualifications - Native speaker of Vietnamese, currently residing in Taiwan or the United States - Fluent in English - Ability to evaluate specific language accents - Quiet environment and laptop/desktop for validation work Requirements - Project-based opportunity with CrowdGen - Join the CrowdGen Community as an Independent Contractor - Receive an email from CrowdGen regarding account creation - Log in to the account and reset the password - Complete the setup requirements and proceed with your application Benefits - Pay: $0.070 USD per unit - Flexible schedule - Work remotely from home

• Write prompts in your native language designed to test an AI model. • Evaluate the AI's responses using a structured rating guide. • Identify areas where the AI's responses can be improved. • Translate your prompts and evaluations into English.

• Write prompts in your native language designed to test an AI model. • Evaluate the AI's responses using a structured rating guide. • Identify areas where the AI's responses can be improved. • Translate your prompts and evaluations into English.

• Write prompts in your native language designed to test an AI model. • Evaluate the AI's responses using a structured rating guide. • Identify areas where the AI's responses can be improved. • Translate your prompts and evaluations into English.

• Write prompts in your native language designed to test an AI model. • Evaluate the AI's responses using a structured rating guide. • Identify areas where the AI's responses can be improved. • Translate your prompts and evaluations into English.

• Review the provided recording guidelines • Record a set of short, scripted voice commands using the Appen Mobile app • Complete a full session of 500 prompts • Ensure at least 90% of your recorded prompts meet the quality standard

• Coordinate assigned project workstreams - covering productivity, qualification, training, crowd communications, and QA - to ensure on-time delivery against execution standards • Deliver high-quality annotation and QA during POC and pilot phases to establish quality benchmarks and de-risk new client relationships from day one • Partner with Client Partners and Pilot Teams to validate quote assumptions, including time-per-task and quality requirements, before finalization • Analyze datasets to surface patterns, insights, and optimization opportunities that inform automation strategy and drive thought leadership content for customers • Support Project Managers and Client Partners as the AI expert point of contact across client meetings, workshops, QBRs, and status updates • Maintain version control and change communication protocols to ensure contributors are informed of guideline updates prior to launch and throughout production • Translate QA findings into targeted improvements to supporting materials, reducing recurring annotation errors and improving overall data quality • Provide data-backed findings and consultancy insight to strengthen Appen's advisory value during critical early project phases and ongoing client relationships

• Listen to audio recordings and create accurate written transcripts of what is spoken. • Transcribing audio recordings into text with high accuracy. • Capturing speech exactly as spoken, including pauses, filler words (e.g., "uh," "um"), repetitions, and incomplete sentences. • Adding timestamps to match the audio. • Identifying and labeling different speakers in conversations. • Marking non-speech events such as laughter, background noise, or unclear audio.