Upload 53 files

.gitattributes

@@ -33,3 +33,12 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+whisper_blas_bin_v1_3_0/emp2.wav filter=lfs diff=lfs merge=lfs -text
+whisper_blas_bin_v1_3_0/libopenblas.dll filter=lfs diff=lfs merge=lfs -text
+whisper_blas_bin_v1_3_0/SDL2.dll filter=lfs diff=lfs merge=lfs -text
+Samples/emp2.wav filter=lfs diff=lfs merge=lfs -text
+Samples/Montreal[[:space:]]Vacation[[:space:]]Travel[[:space:]]Guide[[:space:]][[:space:]]Expedia.mp3 filter=lfs diff=lfs merge=lfs -text
+Samples/Que[[:space:]]es[[:space:]]TED[[:space:]]y[[:space:]]TEDx.mp3 filter=lfs diff=lfs merge=lfs -text
+Samples/test_mixture.wav filter=lfs diff=lfs merge=lfs -text
+Samples/Gujarati_2.wav filter=lfs diff=lfs merge=lfs -text
+Samples/Telugu_2.wav filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,10 +1,10 @@
 ---
 title: Whisper CPP ASR CLI
-emoji: 📊
-colorFrom: purple
+emoji: 🚀
+colorFrom: yellow
 colorTo: pink
 sdk: gradio
-sdk_version: 4.1.1
+sdk_version: 4.0.2
 app_file: app.py
 pinned: false
 license: mit

Samples/Gujarati_2.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d087ac4bc15ddccecbbded4c8fdd2c9501bd8e08c6550f9c03ec46bd2df64da
+size 1597484

Samples/Montreal Vacation Travel Guide Expedia.mp3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc5148bfd0aca841584c6120c9b02c4c3d9aada2e9b015827a7869803b6471cb
+size 8058808

Samples/Que es TED y TEDx.mp3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:120a8f591bec8715dea618a14bf0538ec03444aaf875994bac0d8a1506f2ffdc
+size 1852114

Samples/Telugu_2.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccee4c1d338d4d6fc96dfca5f48ecac3d9b9517e44f0302037fe46304e9e76b0
+size 1122348

Samples/emp2.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:133ac69ac8a7d99f1fe92b7dbd810d7f4e44cd4f282179aa8c5d83183e64ad61
+size 1122066

Samples/test_mixture.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7126269f058724336d28bf6e82fa14b7030de321dbafa48c9fa20884d35dab9d
+size 1546250

app.py

@@ -0,0 +1,79 @@
+import gradio as gr
+import soundfile as sf
+import tempfile
+import shutil
+import os
+import librosa
+import time
+import numpy as np
+import subprocess 
+
+# command = r"""wine './whisper_blas_bin_v1_3_0/main.exe' -h"""
+# wine_command = """sudo apt-get install wine"""
+command2 = """chmod +777 ./whisper_blas_bin_v1_3_0/main.exe"""
+# wine_c = subprocess.run(wine_command,shell=True,stdout=subprocess.PIPE,stderr=subprocess.PIPE, text=True)
+perm = subprocess.run(command2, shell=True,stdout=subprocess.PIPE,stderr=subprocess.PIPE, text=True)
+# result = subprocess.run(command, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
+# print("Wine Installation: ",wine_c)
+print("Access Installation: ",perm)
+# Fpr win32 instalattion while using medium model
+# command3 = "apt install sudo"
+# command4 = "dpkg --add-architecture i386"
+# command5 = "apt-get update"
+# command6 = "apt-get install wine32:i386"
+# t1= subprocess.run(command3, shell=True,stdout=subprocess.PIPE,stderr=subprocess.PIPE, text=True)
+# t2= subprocess.run(command4, shell=True,stdout=subprocess.PIPE,stderr=subprocess.PIPE, text=True)
+# t3= subprocess.run(command5, shell=True,stdout=subprocess.PIPE,stderr=subprocess.PIPE, text=True)
+# t4= subprocess.run(command6, shell=True,stdout=subprocess.PIPE,stderr=subprocess.PIPE, text=True)
+# print("T1: ",t1)
+# print("T2: ",t2)
+# print("T3: ",t3)
+# print("T4: ",t4)
+
+
+def resample_to_16k(audio, orig_sr):
+    y_resampled = librosa.resample(y=audio, orig_sr=orig_sr, target_sr = 16000)
+    return y_resampled
+
+def transcribe(audio):
+    sr,y = audio
+    y = y.astype(np.float32)
+    y /= np.max(np.abs(y))
+    y_resampled = resample_to_16k(y, sr)
+    
+    
+    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as temp_audio:
+        temp_audio_path = temp_audio.name
+        sf.write(temp_audio_path, y_resampled, 16000)
+
+
+    
+    # command = rf"""wine './whisper_blas_bin_v1_3_0/main.exe' -m './whisper_blas_bin_v1_3_0/models/ggml-model-whisper-small.en.bin' -osrt -f '{temp_audio_path}' -nt"""  # English only
+    command = rf"""wine './whisper_blas_bin_v1_3_0/main.exe' -m './whisper_blas_bin_v1_3_0/models/ggml-model-whisper-base.bin' -osrt -f '{temp_audio_path}' -nt"""    # Multilingual
+    # win32 error while using medium model
+    # command = rf"""wine './whisper_blas_bin_v1_3_0/main.exe' -m './whisper_blas_bin_v1_3_0/models/ggml-model-whisper-medium-q5_0.bin' -osrt -f '{temp_audio_path}' -nt"""    # Multilingual
+    
+    start_time = time.time()
+    result = subprocess.run(command, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
+    end_time = time.time()
+    print("Output",result.stdout)
+    print("Error",result.stderr)
+    transcription = result.stdout
+    print(transcription)
+    
+    print("--------------------------")
+    print(f"Execution time: {end_time - start_time} seconds")
+    return transcription, (end_time - start_time)
+
+
+
+demo = gr.Interface(
+    transcribe,
+    inputs = "microphone",
+    # gr.Audio(sources=["microphone"]),
+    outputs = [gr.Textbox(label="CLI_Transcription"),gr.Textbox(label="Time taken for Transcription")],
+    examples=["./Samples/Hindi_1.mp3","./Samples/Hindi_2.mp3","./Samples/Tamil_1.mp3","./Samples/Tamil_2.mp3","./Samples/Marathi_1.mp3","./Samples/Marathi_2.mp3","./Samples/Nepal_1.mp3","./Samples/Nepal_2.mp3","./Samples/Telugu_1.wav","./Samples/Telugu_2.wav","./Samples/Malayalam_1.wav","./Samples/Malayalam_2.wav","./Samples/Gujarati_1.wav","./Samples/Gujarati_2.wav","./Samples/Bengali_1.wav","./Samples/Bengali_2.wav"]
+
+)
+
+demo.launch()

packages.txt

@@ -0,0 +1,4 @@
+sudo
+wine
+libwine
+wine64

requirements.txt

@@ -0,0 +1,2 @@
+librosa
+soundfile

whisper_blas_bin_v1_3_0/SDL2.dll

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6130ad170f5d09a8b3bc6c38a6c7a19f7fd840f345ce0f10fd11bd684bfd1c2
+size 2509824

whisper_blas_bin_v1_3_0/emp2.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:133ac69ac8a7d99f1fe92b7dbd810d7f4e44cd4f282179aa8c5d83183e64ad61
+size 1122066

whisper_blas_bin_v1_3_0/emp2.wav.srt

@@ -0,0 +1,32 @@
+1
+00:00:00,000 --> 00:00:07,120
+ Climate change activists from the last generation movement in Italy climbed into Rome's iconic
+
+2
+00:00:07,120 --> 00:00:11,520
+ Trevi Fountain and used dye to turn its waters black.
+
+3
+00:00:11,520 --> 00:00:15,960
+ They described it as a protest against the inaction of Italy's government and institutions
+
+4
+00:00:15,960 --> 00:00:17,480
+ on climate change.
+
+5
+00:00:17,480 --> 00:00:22,600
+ The group says the continued investment in fossil fuels is directly linked to the catastrophic
+
+6
+00:00:22,600 --> 00:00:27,120
+ flooding that has recently hit Italy's Emilia Ramana region.
+
+7
+00:00:27,120 --> 00:00:33,040
+ This generation's peaceful protests have targeted other cultural sites in the past.
+
+8
+00:00:33,040 --> 00:00:35,620
+ (logo whooshing)
+

whisper_blas_bin_v1_3_0/libopenblas.dll

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7f521b507686ddc25bee7538a80c3741d77f5b7b6c12bddf66894e672eebe0b
+size 48536473

whisper_blas_bin_v1_3_0/models/ggml-model-gpt-2-117M.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b457d5fcc7f2f71e727bee74298d42d80610619e02af16beca53d44a71d5f607
+size 251222425

whisper_blas_bin_v1_3_0/models/ggml-model-whisper-base.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60ed5bc3dd14eea856493d334349b405782ddcaf0028d4b5df4088345fba2efe
+size 147951465

whisper_blas_bin_v1_3_0/models/ggml-model-whisper-base.en.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a03779c86df3323075f5e796cb2ce5029f00ec8869eee3fdfb897afe36c6d002
+size 147964211

whisper_blas_bin_v1_3_0/models/ggml-model-whisper-medium-q5_0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19fea4b380c3a618ec4723c3eef2eb785ffba0d0538cf43f8f235e7b3b34220f
+size 539212467

whisper_blas_bin_v1_3_0/models/ggml-model-whisper-small.en.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6138d6d58ecc8322097e0f987c32f1be8bb0a18532a3f88f734d1bbf9c41e5d
+size 487614201