
Dibuka
Disiarkan
•
Berakhir dalam 22 hari
Dibayar semasa penghantaran
Multilingual Voice Recording Project – Code-Switching Conversations Project Name: BV Project Type: Remote | Ongoing (Limited Slots per Locale) Project Overview Project BV is a multilingual speech data collection initiative designed to enhance Automatic Speech Recognition (ASR) systems for high-value multilingual call center scenarios, including financial services, healthcare, and telecommunications. The project focuses on collecting natural code-switching conversational audio, where speakers alternate between two languages within a single conversational turn. Scripts will be provided; however, natural delivery, fluency, and context-appropriate language switching are essential. Language Requirements Primary Language (Native level – one required): Catalan (ca-ES: Spain) Secondary Language (Required): Catalan or Basque – Fluent Participants may be located anywhere globally, provided they are native in Catalan (ca-ES: Spain) and fluent in English The Project Manager will request language verification samples for both primary and secondary languages. Participation Structure Participation is one-time only per speaker. Two speakers will form a pair (agent and customer roles). Speakers do not need to be in the same location. Scripts and recording instructions will be provided. Zoom is the preferred method for recording the conversations. Audacity is the preferred tool for converting audio files to the required sample rates and for verifying signal-to-noise ratio compliance. Recording Scope Target volume: 200 hours of audio per locale Per pair output: 1.2 hours of recorded audio Quality acceptance: Minimum 80% usable speech hours Compensation (USD – Per Pair, 2 Speakers) All rates below are paid in United States Dollars (USD) and apply to one hour of final audio (30 minutes per speaker). Payment is split evenly between the two participants. Catalan (ca-ES: Spain) USD 40.00 per participant Recording Specifications File Format: Stereo WAV Left channel: Agent Right channel: Customer Sample Rates: 8kHz (telephone quality) – 50% 16kHz – 50% Code-Switching Ratio: 70% primary language 30% English ±10% tolerance Delivery Style: Natural, context-pecific language switching No forced or artificial transitions Speaker Requirements Age Distribution Targets: 18–25 years: 30% 25–40 years: 30% 40–65 years: 20% 65+ years: 20% Gender Balance: 50% male / 50% female Language Proficiency: Fluent Catalan or Basque is mandatory; non-fluent speakers will be rejected Catalan (ca-ES: Spain) is mandatory Participation Limit: Each speaker may participate once only Audio Quality Standards Minimum 25dB signal-to-noise ratio No background noise, distortion, clipping, or echo Quiet, controlled recording environment required Headset or professional microphone strongly recommended How to Apply Applicants should be prepared to provide: * Native language and Catalan or Basque proficiency details * Age range and gender * Confirmation of one-time participation * Language verification samples upon request * Participants will be placed in a group chat with others of the same language to work out who will work with whom for the recordings. The project will close once all required recordings are collected. Apply and wait for the next steps if deemed by our team that you qualify. Note: If you do not hear from us, consider that you do not qualify for this project
ID Projek: 40298990
6 cadangan
Dibuka untuk pembidaan
Projek jarak jauh
Aktif 3 hari yang lalu
Tetapkan bajet dan garis masa anda
Dapatkan bayaran untuk kerja anda
Tuliskan cadangan anda
Ianya percuma untuk mendaftar dan membida pekerjaan
6 pekerja bebas membida secara purata $52 USD untuk pekerjaan ini

I’m based in Catalonia, Spain, and I’m a native Catalan (ca-ES) speaker with fluent English. I’m interested in participating in Project BV and can provide natural, high-quality code-switching conversations as required. I’m comfortable following structured scripts while maintaining a natural and fluent delivery, ensuring smooth transitions between Catalan and English without sounding forced. I understand the importance of context and authenticity, especially for real-world conversational scenarios. I have access to a quiet recording environment and suitable equipment, allowing me to meet the required audio quality standards, including clean sound, proper format and minimal background noise. I’m also familiar with basic recording tools and can follow technical instructions regarding file format, channels and sample rates. I’m available for one-time participation, open to being paired with another speaker, and can provide language verification samples if needed.
$50 USD dalam 7 hari
5.0
5.0

Hello, I will begin by setting up a systematic approach to capturing the code-switching conversations in both Catalan and English. My first step will be to identify key contexts and scenarios relevant to your project, ensuring that the recordings reflect authentic interactions. I will implement a structured framework for the conversation sessions, paying attention to natural flow and linguistic nuances. This will involve careful planning of participant selection and recording environments to generate high-quality audio that meets your project's needs. To ensure a smooth workflow, I’ll establish clear milestones that allow us to evaluate progress incrementally. We can start with a small initial milestone where I provide a brief sample of the recordings. If this aligns with your expectations, we can then proceed with the full scope of the project, allowing for adjustments based on your feedback. Best regards,
$93.80 USD dalam 7 hari
4.3
4.3

I am highly interested in contributing to Project BV and supporting the development of high-quality multilingual ASR systems. I bring over 5 years of experience working on audio-centric and AI-related projects, with a strong focus on data collection, preprocessing, and quality validation. I have worked on conversational speech recording tasks involving structured scripts, ensuring natural delivery and maintaining linguistic balance, including code-switching scenarios. I am proficient with recording workflows using Zoom and audio processing tools like Audacity, including format conversion, stereo channel handling, and ensuring compliance with required sample rates (8kHz/16kHz) and signal-to-noise standards. I consistently maintain high audio clarity with minimal noise and distortion. I understand the importance of natural, context-aware conversations and can deliver smooth, realistic code-switching without sounding forced, while maintaining the required language ratio. I am reliable in coordinating with a recording partner, following detailed instructions, and delivering high-quality output within deadlines, ensuring 80%+ usable speech. I am ready to provide language verification samples and confirm one-time participation. Best regards, Vidhi Mala
$40 USD dalam 1 hari
0.0
0.0

Hello, I am a native Catalan (Spain) and Spanish (Spain) speaker, currently based in Spain, with experience in voice recording and AI speech data projects. I can deliver clear, natural recordings with proper pronunciation, following code-switching requirements between Catalan and English in a realistic conversational style. I have a quiet recording environment and I use Audacity for audio processing, ensuring good signal quality. I can also bring a second speaker if needed, or collaborate with another participant for the paired recordings. I am available to start immediately and follow all project guidelines carefully. Best regards, Francesc.
$42 USD dalam 7 hari
0.0
0.0

I can participate in Project BV as a native-level Catalan (ca-ES) speaker with strong English fluency, and I fully understand the requirements for natural code-switching, structured dialogue delivery, and high-quality ASR data collection. I have experience working with recorded speech tasks, ensuring clean audio capture, consistent pacing, and accurate adherence to scripts while maintaining natural conversational flow. I can comfortably switch between Catalan and English in a realistic, context-aware manner without forced transitions. I can also meet your technical requirements for: • Controlled recording environment (low noise, echo-free) • Proper microphone/headset usage for clean signal quality • Compliance with 8kHz and 16kHz WAV stereo formatting (Left/Right channel structure) • Following SNR and clarity standards for usable speech data I am available for one-time participation and can coordinate smoothly within a paired recording setup via Zoom. I understand the importance of consistency, fluency, and dataset quality for ASR model training. I confirm readiness to provide language verification samples and demographic details as required. What is your preferred timeline for onboarding selected speaker pairs?
$42 USD dalam 7 hari
0.0
0.0

Cessnock, Australia
Kaedah pembayaran disahkan
Ahli sejak Apr 27, 2025
$10-30 USD
$2-8 USD / jam
$465.7 USD
$80-100 USD
min $50 USD / jam
$250-750 USD
£250-750 GBP
£18-36 GBP / jam
$10-5000 USD
$250-750 USD
$250-750 USD
$250-750 USD
$1500-3000 USD
$250-750 USD
$25-50 USD / jam
€250-750 EUR
$42 USD
$250-750 AUD
$10-30 USD
$20-60 USD
$230 AUD
$10-5000 USD
$15-25 USD / jam
$8-15 USD / jam
$30-60 USD / jam