American natural group conversation recording
Group recording, recording is applied to artificial intelligence.
- Must be native American
- Gender: Male:Female = 1:1
- Age: 18-60 years old, young people mainly, with unlimited
- distribution proportion.
- Recording Environment: Relatively quiet environment with no echo.
- Recording Tools I will send you to the play store or app store link
1)The recorder must be of American nationality and standard American pronunciation. Two conditions must be met at the same time. Data with a heavy accent but consistent nationality will not be accepted.
2)The way of speaking is natural, the voice is fluent and clear, and it cannot be read aloud or deliberately simulated.
3)Do not include violence, pornography, political sensitivity, etc. in the content of the conversation.
4) Each recording person participating in the recording cannot speak too little, and a person only has a large number of simple echoes ( " um ", " oh ", and " right " ) as invalid data.
5)When recording, avoid rushing to cause overlapping speech.
6) The content of the recorder’s utterance is consistent with the content of the selected topic, and the duration meets the requirements (the original duration of each topic is in
Within 5-30 minutes )
7) Integrity problem: spectrogram, the waveform chart, recording whether the target environment ( to ensure that no spent on the case, open, current, static segment empty energy, etc. )
8)Speak clearly and earnestly, avoid people with pathological pronunciation such as stuttering, big tongue, hoarse voice, etc.
9)Participants ( or their guardians ) must understand the recording requirements ( recording method, recording environment, reading requirements, qualification indicators ) in advance.
10)The recording person or his guardian needs to sign an Apps.
11)Each person can only record one recording of the project. If one person records multiple copies of data, the corresponding compensation will be canceled.