describing metadata
Browse files
README.md
CHANGED
@@ -67,6 +67,25 @@ configs:
|
|
67 |
A new publicly available dataset consisting of 289 life story interviews (365 hours), featuring a broad range of speakers varying in age, education, and regional accents.
|
68 |
|
69 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
70 |
## Dataset
|
71 |
| Hugging Face |
|
72 |
| ------------ |
|
|
|
67 |
A new publicly available dataset consisting of 289 life story interviews (365 hours), featuring a broad range of speakers varying in age, education, and regional accents.
|
68 |
|
69 |
|
70 |
+
## Metadata:
|
71 |
+
- **audio_id**: Sequential id for the interview;
|
72 |
+
- **audio_name**: Unique code for the interview;
|
73 |
+
- **file_path**: Wav audio file path;
|
74 |
+
- **speaker_type**: R is the interviewee, P/1 is interviewer 1, P/2 is interviewer 2 and so on;
|
75 |
+
- **speaker_code**: Unique code for the speaker;
|
76 |
+
- **speaker_gender**: Gender of the speaker;
|
77 |
+
- **education**: Education level of the interviewee, filled only when speaker_type = 'R';
|
78 |
+
- **birth_state**: Birth state (region) of the interviewee, filled only when speaker_type = 'R';
|
79 |
+
- **birth_country**: Birth country of the speaker;
|
80 |
+
- **age**: Age of the interviewee, filled only when speaker_type = 'R';
|
81 |
+
- **recording_year**: The year when the audio was recorded;
|
82 |
+
- **audio_quality**: Can be high or low;
|
83 |
+
- **start_time**: The start time in the original complete audio file;
|
84 |
+
- **end_time**: The end time in the original complete audio file;
|
85 |
+
- **duration**: The duration of the segment;
|
86 |
+
- **normalized_text**: Text normalized in lowercase and without punctuation marks;
|
87 |
+
- **original_text**: Text before normalization.
|
88 |
+
|
89 |
## Dataset
|
90 |
| Hugging Face |
|
91 |
| ------------ |
|