1. Voice record

2. copy

*****************

The targeted languages:

Chinese

Japanese

– Korean

Dutch

*****************

Project Description:

The requirements are:

– 5 hours Studio Quality speech reading for each loudspeaker

– 2 speakers for each language (1 male / 1 female)

– High -resolution text texts (without audio) for all sound

– The sound is delivered as WAV files 48 kHz / 32 Tat

___

Registered voice in the processed audio environment:

– Noise floor <-60DB

– RT660 frequency <90MS

___

– Raw sound (there is no cancellation / noise treatment)

Voice Choose: The seller must provide 8 final options for each sound

– Contemporary program content: Conversation, Assistant/Chatbot field

*****************

Data set size:

Total: 40 hours of data

– 10 hours for each language

*****************

Questions for sellers (please include in your response):

1.

2. Can you meet the required sound specifications (noise floor <-60DB, RT60 <90MS, 48 kHz / 32-bit Mono WAF)?

3. Do you have access to professional studio facilities?

4. How many sample sounds for each sex for each language you can offer?

5. Can you save up to 8 final options for each sex if necessary?

6. What is the time for the expected transformation once the final sounds are chosen?

7. Does your prices include texts, or is this separately quoted?

Note:
Please send you again to us. After the internal review, we will provide you with guidance principles of the sample.

Vote

(Disappeared)

To fill
(Disappeared)