**Dataset Description (5–8 words):** French audio clips needing phoneme transcripts **Data Type:** Audio **Subject Matter/Industry (5–8 words):** French speech phonetics and ASR training **Pre-labeled Data:** No **Labeling Software:** Other **Label Types (multi-select):** * Data Collection * Classification **Labeling Overview:** **Qualifications:** You must be native-level in French and comfortable producing or validating ARPABET phonemic transcriptions at an intermediate to expert level. You should be able to reliably detect reading errors and speech disfluencies with high accuracy. Backgrounds in phonetics/linguistics, ASR datasets, or child-speech transcription are strongly preferred. You must be able to start within one week, pass a qualification quiz, and commit 20+ hours per week at the published rate. **What you’ll do:** You will convert large batches of long (1–2 minute) French audio clips into precise word- and phoneme-level transcripts using our web-based annotation tool. Work is delivered in rolling batches, and your annotations will go through a peer QA loop before final acceptance. This work will directly support next-generation speech recognition and language-learning models. Payment is per approved task—higher throughput and accuracy mean higher earnings. **Required Locations:** Global - Any Location **Required English Level:** Conversational **Other Qualifications & Requirements (5–10 bullets):** * Native-level French (able to judge fine-grained pronunciation and reading accuracy). * Intermediate to expert proficiency with **ARPABET** phonemic transcription for French speech. * Demonstrated ability to identify and label **reading errors** (misread words, substitutions, omissions, insertions). * Demonstrated ability to identify and label **disfluencies** (filled pauses, repetitions, false starts, self-corrections). * Prior experience with at least one: phonetics/linguistics work, ASR dataset transcription, phoneme-level labeling, or child-speech transcription. * Able to produce **word-level alignment** and **phoneme-level transcription** with high consistency and attention to detail. * Must be able to start within **1 week** of acceptance. * Must pass a **qualification quiz** before receiving production batches. * Availability to commit **20+ hours/week** on an ongoing basis. * Comfortable working in a **web-based annotation tool** and following peer-QA feedback loops.
Total Budget
$2,200
Pay per Label
$22/hr
Time Requirement
20+ hrs/week
Duration
3-6 months
French audio clips needing phoneme transcripts
Software
Hiring Type
Required Location
Workload / Schedule
20 hours per week, Flexible and able to start immediately
Software
Data Type
Label Types
Subject Matter / Industry
Language
Job Type
Share link