Native Burmese Dialogue Recordings
Budget: ₹600 – ₹1,500 INR
I need native Burmese speakers to read and record a set of short, everyday-conversation sentences. The purpose is to build a high-quality audio corpus for training an automatic speech-recognition system, so clarity is everything.
Please note: Hiring and Communication will be done exclusively on Freelancer only.
Each participant can record up to 1,000 sentences. Due to project requirements, we will provide prompts. You can first think of a sentence based on the field information and keywords in each prompt, and then speak it naturally.
Please record in a quiet room so there is no background noise, echo, or overlapping voices. A modern smartphone or a good USB microphone connected to a computer is fine as long as the audio is clean and consistent.
Deliverables:
1) Think of complete sentences according to the prompt provided. Do not say incomplete sentences or just one phrase.
2) Single audio content requires complete semantic meaning and correct grammar. Stuttering, repeated reading, stuttering and other phenomena are strictly prohibited.
3) Each recording must be between 4 and 9 seconds long. It cannot be shorter than 4 seconds or longer than 9 seconds.
4) One clear audio per sentence
5) Correct pronunciation, natural intonation, no clipping or distortion
Let me know if you have any questions about the recording setup or file format before you start, and I’ll be happy to clarify.
Please note: Hiring and Communication will be done exclusively on Freelancer only.
Each participant can record up to 1,000 sentences. Due to project requirements, we will provide prompts. You can first think of a sentence based on the field information and keywords in each prompt, and then speak it naturally.
Please record in a quiet room so there is no background noise, echo, or overlapping voices. A modern smartphone or a good USB microphone connected to a computer is fine as long as the audio is clean and consistent.
Deliverables:
1) Think of complete sentences according to the prompt provided. Do not say incomplete sentences or just one phrase.
2) Single audio content requires complete semantic meaning and correct grammar. Stuttering, repeated reading, stuttering and other phenomena are strictly prohibited.
3) Each recording must be between 4 and 9 seconds long. It cannot be shorter than 4 seconds or longer than 9 seconds.
4) One clear audio per sentence
5) Correct pronunciation, natural intonation, no clipping or distortion
Let me know if you have any questions about the recording setup or file format before you start, and I’ll be happy to clarify.