Spaces:
Sleeping
Sleeping
Aditi Tewari
commited on
Commit
·
b77c785
1
Parent(s):
b7fe9cf
Update backend and frontend
Browse files- soundscripter.html +2 -2
- soundscripter_flaskAPI.py +7 -5
soundscripter.html
CHANGED
@@ -16,7 +16,7 @@
|
|
16 |
|
17 |
<nav class="navbar" style="background-color:whitesmoke; height:98px; border-radius:30px" >
|
18 |
<div>
|
19 |
-
<img src="wave.png" alt="SoundScripter" width="60" height="70" style="margin-top:-5px">
|
20 |
<span class="navbar-brand" style="font-size:35px;"><b>SoundScripter</b></span>
|
21 |
</div>
|
22 |
<h5>Automatic Speech Recognition</h5>
|
@@ -117,7 +117,7 @@
|
|
117 |
})
|
118 |
.then((response) => response.json())
|
119 |
.then((data) => {
|
120 |
-
if (data.
|
121 |
outputText.value = data.text;
|
122 |
} else {
|
123 |
outputText.value = "No text recognized.";
|
|
|
16 |
|
17 |
<nav class="navbar" style="background-color:whitesmoke; height:98px; border-radius:30px" >
|
18 |
<div>
|
19 |
+
<img src="https://huggingface.co/spaces/adititewari/SoundScripter/blob/main/wave.png" alt="SoundScripter" width="60" height="70" style="margin-top:-5px">
|
20 |
<span class="navbar-brand" style="font-size:35px;"><b>SoundScripter</b></span>
|
21 |
</div>
|
22 |
<h5>Automatic Speech Recognition</h5>
|
|
|
117 |
})
|
118 |
.then((response) => response.json())
|
119 |
.then((data) => {
|
120 |
+
if (data.text) {
|
121 |
outputText.value = data.text;
|
122 |
} else {
|
123 |
outputText.value = "No text recognized.";
|
soundscripter_flaskAPI.py
CHANGED
@@ -30,7 +30,8 @@ async def read_root():
|
|
30 |
# Provide the path to the HTML file containing the front-end code
|
31 |
with open("soundscripter.html", "r") as file:
|
32 |
html_content = file.read()
|
33 |
-
return HTMLResponse(content=html_content)
|
|
|
34 |
|
35 |
|
36 |
def convert_audio_format(input_data, input_format, output_format='wav'):
|
@@ -52,11 +53,11 @@ def recognize_speech(audio_data, language="hi-IN"):
|
|
52 |
return f"API request failed: {e}"
|
53 |
|
54 |
@app.post("/asr")
|
55 |
-
async def transcribe_audio(
|
56 |
-
contents = await
|
57 |
|
58 |
# Determine the input audio format (assumes the format is part of the file name)
|
59 |
-
input_format =
|
60 |
|
61 |
# Convert audio to WAV format
|
62 |
wav_data = convert_audio_format(contents, input_format)
|
@@ -68,7 +69,8 @@ async def transcribe_audio(file: UploadFile = File(...)):
|
|
68 |
|
69 |
# Transcribe the audio
|
70 |
result = recognize_speech(wav_data)
|
71 |
-
|
|
|
72 |
# return {"Text": result}
|
73 |
return JSONResponse(content={"text": result})
|
74 |
|
|
|
30 |
# Provide the path to the HTML file containing the front-end code
|
31 |
with open("soundscripter.html", "r") as file:
|
32 |
html_content = file.read()
|
33 |
+
# return HTMLResponse(content=html_content)
|
34 |
+
return html_content
|
35 |
|
36 |
|
37 |
def convert_audio_format(input_data, input_format, output_format='wav'):
|
|
|
53 |
return f"API request failed: {e}"
|
54 |
|
55 |
@app.post("/asr")
|
56 |
+
async def transcribe_audio(audio: UploadFile = File(...)):
|
57 |
+
contents = await audio.read()
|
58 |
|
59 |
# Determine the input audio format (assumes the format is part of the file name)
|
60 |
+
input_format = audio.filename.split('.')[-1].lower()
|
61 |
|
62 |
# Convert audio to WAV format
|
63 |
wav_data = convert_audio_format(contents, input_format)
|
|
|
69 |
|
70 |
# Transcribe the audio
|
71 |
result = recognize_speech(wav_data)
|
72 |
+
print(result)
|
73 |
+
# print(JSONResponse(content={"text": result}))
|
74 |
# return {"Text": result}
|
75 |
return JSONResponse(content={"text": result})
|
76 |
|