Upload original metadata: original_metadata/README.txt

original_metadata/README.txt

@@ -0,0 +1,130 @@
+The Expresso Dataset
+====================
+
+Introduction
+------------
+
+We introduce Expresso, a high-quality (48kHz) expressive speech dataset that includes both expressively rendered read speech (8 styles, in mono wav format) and improvised dialogues (26 styles, in stereo wav format). The dataset includes 4 speakers (2 males, 2 females), and totals 40 hours (11h read, 30h improvised). The transcriptions of the read speech are also provided.
+
+
+Data Statistics
+---------------
+
+Here is the statistics of Expresso's expressive styles:
+
+----------------------------------------------------------------
+ Style            | Read (min) | Improvised (min) | total (hrs)
+------------------|------------|------------------|-------------
+ angry            | -          | 82               | 1.4
+ animal           | -          | 27               | 0.4
+ animal_directed  | -          | 32               | 0.5
+ awe              | -          | 92               | 1.5
+ bored            | -          | 92               | 1.5
+ calm             | -          | 93               | 1.6
+ child            | -          | 28               | 0.4
+ child_directed   | -          | 38               | 0.6
+ confused         | 94         | 66               | 2.7
+ default          | 133        | 158              | 4.9
+ desire           | -          | 92               | 1.5
+ disgusted        | -          | 118              | 2.0
+ enunciated       | 116        | 62               | 3.0
+ fast             | -          | 98               | 1.6
+ fearful          | -          | 98               | 1.6
+ happy            | 74         | 92               | 2.8
+ laughing         | 94         | 103              | 3.3
+ narration        | 21         | 76               | 1.6
+ non_verbal       | -          | 32               | 0.5
+ projected        | -          | 94               | 1.6
+ sad              | 81         | 101              | 3.0
+ sarcastic        | -          | 106              | 1.8
+ singing*         | -          | 4                | .07
+ sleepy           | -          | 93               | 1.5
+ sympathetic      | -          | 100              | 1.7
+ whisper          | 79         | 86               | 2.8
+ **Total**        | **11.5h*   | **34.4h**        | **45.9h**
+----------------------------------------------------------------
+*singing is the only improvised style that is not in dialogue format.
+
+
+Audio Quality
+-------------
+
+The audio was recorded in a professional recording studio with minimal background noise at 48kHz/24bit. The files for read speech and singing are in a mono wav format; and for the dialog section in stereo (one channel per actor), where the original flow of turn-taking is preserved.
+
+
+Directory Structure
+-------------------
+
+The expresso dataset directory has the following structure:
+```
+expresso/
+|---README.txt
+|---LICENSE.txt
+|---read_transcriptions.txt
+|---VAD_segments.txt
+|---splits/
+|   |---train.txt
+|   |---dev.txt
+|   |---test.txt
+|   `---README
+`---audio_48khz/
+    |---conversational/
+    |   |---ex04-ex01/ # speaker pair in {channel1_spk}-{channel2_spk} format
+    |   |   |---animal-animaldir/ # style pair in {channel1_style}-{channel2_style} format
+    |   |   |   |---ex04-ex01_animal-animaldir_005.wav
+    |   |   |   |---ex04-ex01_animal-animaldir_006.wav
+    |   |   |   `---...
+    |   |   |---laughing/ # both channels have the same style
+    |   |   |   |---ex04-ex01_laughing_001.wav
+    |   |   |   |---ex04-ex01_laughing_002.wav
+    |   |   |   `---...
+    |   |   `---...
+    |   `---...
+    `---read/
+        |---ex03/ # speaker
+        |   |---default/ # style
+        |   |   |---longform/ # recorded in long format
+        |   |   |   `---ex03_default_longform_00001.wav
+        |   |   `---base/ # recorded in short sentences
+        |   |       |---ex03_default_00003.wav
+        |   |       |---ex03_default_emphasis_00010.wav
+        |   |       |---ex03_default_essentials_00005.wav
+        |   |       `---...
+        |   |---happy/
+        |   |   `---base/
+        |   |       |---ex04_happy_00085.wav
+        |   |       |---ex04_happy_00091.wav
+        |   |       `---...
+        |   `---...
+        `---...
+```
+
+
+Audio Naming Convention
+-----------------------
+
+The "conversational" audio directory contains the improvised dialogues (except singing) and is organized with the following structure:
+`conversational/{speaker pair}/{styles}/{speaker pair}_{styles}_{id}.wav`
+where
+- {speaker pair}: hyphen-separated pair of speakers in two channels, with the left speaker on first/left channel, and the right speaker on second/right channel. E.g. "ex03-ex01" means "ex03" on channel 1 and "ex01" on channel 2
+- {styles}: styles of the two channels of the dialogue. If {styles} contains a hyphen, the scene contains two styles. The style on the left corresponds to the channel1/left speaker�s style, and the style on the right corresponds to the channel2/right speaker�s style. If the is no hyphen, both speakers are speaking in the same style
+- {id}: ID of the audio with the given speaker pair and style
+
+The "read" audio directory contains all the read speech and singing style, with the following structure:
+`read/{speaker}/{style}/{corpus}/{speaker}_{style or substyle}_{id}.wav`
+where
+- {speaker}: speaker name
+- {style}: expressive style of the audio
+- {corpus}: 'base' or 'longform'
+- {style or substyle}: same as {style}, with the exception of longform speech and default substyles. The exception substyles are: 'narration_longform', 'default_longform', 'default_emphasis', 'default_essentials'
+- {id}: ID of the audio with the given speaker and style/substyle
+
+
+Licence
+-------
+The Expresso dataset is distributed under the [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/) license.
+
+
+Reference
+---------
+For more information, see the paper "EXPRESSO: A Benchmark and Analysis of Discrete Expressive Speech Resynthesis", Tu Anh Nguyen*, Wei-Ning Hsu*, Antony D'Avirro*, Bowen Shi*, Itai Gat, Maryam Fazel-Zarani, Tal Remez, Jade Copet, Gabriel Synnaeve, Michael Hassid, Felix Kreuk, Yossi Adi+, Emmanuel Dupoux+, INTERSPEECH 2023.