Datasets:

Cnam-LMSSC
/

vibravox

@@ -64,10 +64,10 @@ vibravox = load_dataset("Cnam-LMSSC/vibravox", subset, streaming=True)
 - **Homepage:** For more information about the project, visit our project page on [https://vibravox.cnam.fr](https://vibravox.cnam.fr)
-- **Github repository:** [jhauret/vibravox](https://github.com/jhauret/vibravox) : Source code for Automatic Speech Recognition, Bandwidth Extension and Speaker Verification using the Vibravox dataset
 - **Point of Contact:** [Eric Bavu](https://acoustique.cnam.fr/contacts/bavu/en/#contact)
-- **Curated by:** AVA Team of the [LMSSC Research Laboratory](https://lmssc.cnam.fr)
-- **Funded by:** French [Agence Nationale Pour la Recherche / AHEAD Project](https://anr.fr/en/funded-projects-and-impact/funded-projects/project/funded/project/b2d9d3668f92a3b9fbbf7866072501ef-5aac4914c7/?tx_anrprojects_funded%5Bcontroller%5D=Funded&cHash=fa352121b44b60bf6a5917180d5205e6)
 - **Language:** French
 - **License:** Creative Commons Attributions 4.0
@@ -192,7 +192,7 @@ The 4 subsets correspond to :
 ### Splits
-All the subsets are available in 3 splits (train, validation and test), with a standard 80 / 10 / 10 repartition, without overlapping any speaker in each split.
 The speakers / participants in specific splits are the same for each subset, thus allowing to
@@ -207,10 +207,10 @@ In non-streaming mode (default), the path value of all dataset.Audio dictionnary
 * `audio.headset_mic` (datasets.Audio) - a dictionary containing the path to the audio recorded by the headset microphone, the decoded (mono) audio array, and the sampling rate.
 * `audio.forehead_accelerometer` (datasets.Audio) - a dictionary containing the path to the audio recorded by the forehead miniature accelerometer, the decoded (mono) audio array, and the sampling rate.
-*  `audio.soft_in_ear_mic` (datasets.Audio) - a dictionary containing the path to the audio recorded by the in-ear soft foam-embedded microphone, the decoded (mono) audio array, and the sampling rate.
-*  `audio.rigid_in_ear_mic` (datasets.Audio) - a dictionary containing the path to the audio recorded by the in-ear rigid earpiece-embedded microphone, the decoded (mono) audio array, and the sampling rate.
-*  `audio.temple_vibration_pickup` (datasets.Audio) - a dictionary containing the path to the audio recorded by the temple vibration pickup, the decoded (mono) audio array, and the sampling rate.
-*  `audio.laryngophone` (datasets.Audio) - a dictionary containing the path to the audio recorded by the piezeoelectric laryngophone, the decoded (mono) audio array, and the sampling rate.
 * `gender` (string) - gender of speaker (```male```or ```female```)
 * `speaker_id` (string) - encrypted id of speaker
 * `duration` (float32) - the audio length in seconds.
@@ -385,7 +385,7 @@ TODO : update values when final dataset is uploaded
 ### Textual source data
-The text read by all participants is collected from the French Wikipedia subset of Common voice ( [link1](https://github.com/common-voice/common-voice/blob/main/server/data/fr/wiki-1.fr.txt) [link2](https://github.com/common-voice/common-voice/blob/main/server/data/fr/wiki-2.fr.txt) )  . We applied some additional filters to these textual datasets in order to create a simplified dataset with a minimum number of tokens and to reduce the uncertainty of the pronunciation of some proper names. We therefore removed all proper names except common first names and the list of french towns. We also removed any utterances that contain numbers, Greek letters, math symbols, or that are syntactically incorrect.
 All lines of the textual source data from Wikipedia-extracted textual dataset has then been phonemized using  the [bootphon/phonemizer](https://github.com/bootphon/phonemizer) and manually edited to only keep strict french IPA characters.

 - **Homepage:** For more information about the project, visit our project page on [https://vibravox.cnam.fr](https://vibravox.cnam.fr)
+- **Github repository:** [jhauret/vibravox](https://github.com/jhauret/vibravox) : Source code for ASR, BWE and SPKV tasks using the Vibravox dataset
 - **Point of Contact:** [Eric Bavu](https://acoustique.cnam.fr/contacts/bavu/en/#contact)
+- **Curated by:** [AVA Team](https://lmssc.cnam.fr/fr/recherche/identification-localisation-synthese-de-sources-acoustiques-et-vibratoires) of the [LMSSC Research Laboratory](https://lmssc.cnam.fr)
+- **Funded by:** [Agence Nationale Pour la Recherche / AHEAD Project](https://anr.fr/en/funded-projects-and-impact/funded-projects/project/funded/project/b2d9d3668f92a3b9fbbf7866072501ef-5aac4914c7/?tx_anrprojects_funded%5Bcontroller%5D=Funded&cHash=fa352121b44b60bf6a5917180d5205e6)
 - **Language:** French
 - **License:** Creative Commons Attributions 4.0
 ### Splits
+All the subsets are available in 3 splits (train, validation and test), with a standard 80% / 10% / 10% repartition, without overlapping any speaker in each split.
 The speakers / participants in specific splits are the same for each subset, thus allowing to
 * `audio.headset_mic` (datasets.Audio) - a dictionary containing the path to the audio recorded by the headset microphone, the decoded (mono) audio array, and the sampling rate.
 * `audio.forehead_accelerometer` (datasets.Audio) - a dictionary containing the path to the audio recorded by the forehead miniature accelerometer, the decoded (mono) audio array, and the sampling rate.
+* `audio.soft_in_ear_mic` (datasets.Audio) - a dictionary containing the path to the audio recorded by the in-ear soft foam-embedded microphone, the decoded (mono) audio array, and the sampling rate.
+* `audio.rigid_in_ear_mic` (datasets.Audio) - a dictionary containing the path to the audio recorded by the in-ear rigid earpiece-embedded microphone, the decoded (mono) audio array, and the sampling rate.
+* `audio.temple_vibration_pickup` (datasets.Audio) - a dictionary containing the path to the audio recorded by the temple vibration pickup, the decoded (mono) audio array, and the sampling rate.
+* `audio.laryngophone` (datasets.Audio) - a dictionary containing the path to the audio recorded by the piezeoelectric laryngophone, the decoded (mono) audio array, and the sampling rate.
 * `gender` (string) - gender of speaker (```male```or ```female```)
 * `speaker_id` (string) - encrypted id of speaker
 * `duration` (float32) - the audio length in seconds.
 ### Textual source data
+The text read by all participants is collected from the French Wikipedia subset of Common voice ( [link1](https://github.com/common-voice/common-voice/blob/6e43e7e61318bf4605b59379e3f35ba5333d7a29/server/data/fr/wiki-1.fr.txt) [link2](https://github.com/common-voice/common-voice/blob/6e43e7e61318bf4605b59379e3f35ba5333d7a29/server/data/fr/wiki-2.fr.txt) )  . We applied some additional filters to these textual datasets in order to create a simplified dataset with a minimum number of tokens and to reduce the uncertainty of the pronunciation of some proper names. We therefore removed all proper names except common first names and the list of french towns. We also removed any utterances that contain numbers, Greek letters, math symbols, or that are syntactically incorrect.
 All lines of the textual source data from Wikipedia-extracted textual dataset has then been phonemized using  the [bootphon/phonemizer](https://github.com/bootphon/phonemizer) and manually edited to only keep strict french IPA characters.