From dc2fd900616aef02248313fb29e9e3be5bb25f1c Mon Sep 17 00:00:00 2001
From: Hironsan <light.tree.1.13@gmail.com>
Date: Tue, 18 May 2021 14:34:22 +0900
Subject: [PATCH] Add upload catalog for speech to text

---
 backend/api/views/upload/catalog.py  | 10 +++++++++-
 backend/api/views/upload/examples.py |  6 ++++++
 2 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/backend/api/views/upload/catalog.py b/backend/api/views/upload/catalog.py
index 82e91029..57b89d8c 100644
--- a/backend/api/views/upload/catalog.py
+++ b/backend/api/views/upload/catalog.py
@@ -5,7 +5,7 @@ from pydantic import BaseModel
 from typing_extensions import Literal
 
 from ...models import (DOCUMENT_CLASSIFICATION, IMAGE_CLASSIFICATION, SEQ2SEQ,
-                       SEQUENCE_LABELING)
+                       SEQUENCE_LABELING, SPEECH2TEXT)
 from . import examples
 
 encodings = Literal[
@@ -168,6 +168,11 @@ class ImageFile(Format):
     accept_types = 'image/png, image/jpeg, image/bmp, image/gif'
 
 
+class AudioFile(Format):
+    name = 'AudioFile'
+    accept_types = 'audio/ogg, audio/aac, audio/mpeg, audio/wav'
+
+
 class OptionColumn(BaseModel):
     encoding: encodings = 'utf_8'
     column_data: str = 'text'
@@ -241,3 +246,6 @@ Options.register(SEQ2SEQ, Excel, OptionColumn, examples.Text_CSV)
 
 # Image classification
 Options.register(IMAGE_CLASSIFICATION, ImageFile, OptionNone, examples.Generic_ImageFile)
+
+# Speech to Text
+Options.register(SPEECH2TEXT, AudioFile, OptionNone, examples.Generic_AudioFile)
diff --git a/backend/api/views/upload/examples.py b/backend/api/views/upload/examples.py
index 0b738252..1dbc337c 100644
--- a/backend/api/views/upload/examples.py
+++ b/backend/api/views/upload/examples.py
@@ -10,6 +10,12 @@ Generic_ImageFile = """
 └── 202104210945.png
 """
 
+Generic_AudioFile = """
+├── 202104210943.mp3
+├── 202104210944.mp3
+└── 202104210945.mp3
+"""
+
 Generic_TextLine = """
 Terrible customer service.
 Really great transaction.