Title of the paper: Analysis of conversational speech with application to voice adaptation
(Communicated to ASRU 2021)
System 2 is not included since it fails to generate the original speaker's voice
Generated audios are the hindi translations of the target audios.
System 2 is not included since it fails to generate the original speaker's voice
(Communicated to ASRU 2021)
Experiment 1 : Monolingual Voice Adaptation
| Examples | Systems | |||
|---|---|---|---|---|
| Original | System 1 | System 3 | System 4 | |
| Example 1 |
|
|
|
|
| Example 2 |
|
|
|
|
| Example 3 |
|
|
|
|
| Example 4 |
|
|
|
|
| Example 5 |
|
|
|
|
System 2 is not included since it fails to generate the original speaker's voice
Experiment 2 - Cross-Lingual Voice Adaptation
| Examples | Systems | |||
|---|---|---|---|---|
| Original(English) | System 1 | System 3 | System 4 | |
| Example 1 |
|
|
|
|
| Example 2 |
|
|
|
|
| Example 3 |
|
|
|
|
| Example 4 |
|
|
|
|
| Example 5 |
|
|
|
|
System 2 is not included since it fails to generate the original speaker's voice