Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -31,11 +31,14 @@ def chunks_to_srt(chunks):
|
|
31 |
srt_format += f"{i}\n{start_time_hms} --> {end_time_hms}\n{chunk['text']}\n\n"
|
32 |
return srt_format
|
33 |
|
34 |
-
def transcribe(inputs, task, return_timestamps):
|
35 |
if inputs is None:
|
36 |
raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
|
37 |
-
|
38 |
-
|
|
|
|
|
|
|
39 |
|
40 |
if return_timestamps:
|
41 |
return chunks_to_srt(result['chunks'])
|
@@ -135,6 +138,7 @@ file_transcribe = gr.Interface(
|
|
135 |
gr.inputs.Audio(source="upload", type="filepath", optional=True, label="Audio file"),
|
136 |
gr.inputs.Radio(["transcribe", "translate"], label="Task", default="transcribe"),
|
137 |
gr.inputs.Checkbox(label="Return timestamps"),
|
|
|
138 |
],
|
139 |
outputs="text",
|
140 |
layout="horizontal",
|
@@ -154,6 +158,7 @@ yt_transcribe = gr.Interface(
|
|
154 |
gr.inputs.Textbox(lines=1, placeholder="Paste the URL to a YouTube video here", label="YouTube URL"),
|
155 |
gr.inputs.Radio(["transcribe", "translate"], label="Task", default="transcribe"),
|
156 |
gr.inputs.Checkbox(label="Return timestamps"),
|
|
|
157 |
],
|
158 |
outputs=["html", "text"],
|
159 |
layout="horizontal",
|
|
|
31 |
srt_format += f"{i}\n{start_time_hms} --> {end_time_hms}\n{chunk['text']}\n\n"
|
32 |
return srt_format
|
33 |
|
34 |
+
def transcribe(inputs, task, language, return_timestamps):
|
35 |
if inputs is None:
|
36 |
raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
|
37 |
+
|
38 |
+
# Map the language names to their corresponding codes
|
39 |
+
language_codes = {"English": "en", "Korean": "ko", "Japanese": "ja"}
|
40 |
+
language_code = language_codes.get(language, "en") # Default to "en" if the language is not found
|
41 |
+
result = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={"task": task, "language": f"<|{language_code}|>"}, return_timestamps=return_timestamps)
|
42 |
|
43 |
if return_timestamps:
|
44 |
return chunks_to_srt(result['chunks'])
|
|
|
138 |
gr.inputs.Audio(source="upload", type="filepath", optional=True, label="Audio file"),
|
139 |
gr.inputs.Radio(["transcribe", "translate"], label="Task", default="transcribe"),
|
140 |
gr.inputs.Checkbox(label="Return timestamps"),
|
141 |
+
gr.inputs.Dropdown(choices=["English", "Korean", "Japanese"], label="Language"),
|
142 |
],
|
143 |
outputs="text",
|
144 |
layout="horizontal",
|
|
|
158 |
gr.inputs.Textbox(lines=1, placeholder="Paste the URL to a YouTube video here", label="YouTube URL"),
|
159 |
gr.inputs.Radio(["transcribe", "translate"], label="Task", default="transcribe"),
|
160 |
gr.inputs.Checkbox(label="Return timestamps"),
|
161 |
+
gr.inputs.Dropdown(choices=["English", "Korean", "Japanese"], label="Language"),
|
162 |
],
|
163 |
outputs=["html", "text"],
|
164 |
layout="horizontal",
|