Merge pull request #50 from NavodPeiris/dev

add flexible dependency versioning

.gitignore

@@ -2,4 +2,12 @@ venv
 build
 dist
 speechlib.egg-info
-.env
+.env
+
+*.swp
+*.swo
+
+# By default do not include these files for version control
+# Override this by using 'git add -f'
+*.wav
+*.mp3

README.md

@@ -69,7 +69,9 @@ Transcriptor method takes 7 arguments.
 
 4. model size ("tiny", "small", "medium", "large", "large-v1", "large-v2", "large-v3")
 
-5. ACCESS_TOKEN: huggingface acccess token (also get permission to access `pyannote/[email protected]`)
+5. ACCESS_TOKEN: huggingface acccess token
+    1. Permission to access `pyannote/[email protected]` and `pyannote/segmentation`
+    2. Token requires permission for 'Read access to contents of all public gated repos you can access'
 
 6. voices_folder (contains speaker voice samples for speaker recognition)
 
@@ -86,13 +88,16 @@ transcript will also indicate the timeframe in seconds where each speaker speaks
 ### Transcription example:
 
 ```
+import os
+from speechlib import Transcriptor
+
 file = "obama_zach.wav"  # your audio file
 voices_folder = "" # voices folder containing voice samples for recognition
 language = "en"          # language code
 log_folder = "logs"      # log folder for storing transcripts
 modelSize = "tiny"     # size of model to be used [tiny, small, medium, large-v1, large-v2, large-v3]
 quantization = False   # setting this 'True' may speed up the process but lower the accuracy
-ACCESS_TOKEN = "your hf key" # get permission to access pyannote/[email protected] on huggingface
+ACCESS_TOKEN = "huggingface api key" # get permission to access pyannote/[email protected] on huggingface
 
 # quantization only works on faster-whisper
 transcriptor = Transcriptor(file, log_folder, language, modelSize, ACCESS_TOKEN, voices_folder, quantization)
@@ -110,7 +115,7 @@ res = transcriptor.custom_whisper("D:/whisper_tiny_model/tiny.pt")
 res = transcriptor.huggingface_model("Jingmiao/whisper-small-chinese_base")
 
 # use assembly ai model
-res = transcriptor.assemby_ai_model("your api key")
+res = transcriptor.assemby_ai_model("assemblyAI api key")
 
 res --> [["start", "end", "text", "speaker"], ["start", "end", "text", "speaker"]...]
 ```
@@ -211,4 +216,4 @@ This library uses following huggingface models:
 
 #### https://huggingface.co/speechbrain/spkrec-ecapa-voxceleb
 #### https://huggingface.co/Ransaka/whisper-tiny-sinhala-20k-8k-steps-v2
-#### https://huggingface.co/pyannote/speaker-diarization
+#### https://huggingface.co/pyannote/speaker-diarization

examples/.gitignore

@@ -9,4 +9,5 @@ greek_convo_short.mp3
 greek_convo_short.wav
 my_test.py
 greek_convo.mp3
-greek_convo.wav
+greek_convo.wav
+.env

examples/transcribe.py

@@ -1,3 +1,4 @@
+import os
 from speechlib import Transcriptor
 
 file = "obama_zach.wav"  # your audio file
@@ -6,7 +7,7 @@
 log_folder = "logs"      # log folder for storing transcripts
 modelSize = "tiny"     # size of model to be used [tiny, small, medium, large-v1, large-v2, large-v3]
 quantization = False   # setting this 'True' may speed up the process but lower the accuracy
-ACCESS_TOKEN = "your hf key" # get permission to access pyannote/[email protected] on huggingface
+ACCESS_TOKEN = "huggingface api key" # get permission to access pyannote/[email protected] on huggingface
 
 # quantization only works on faster-whisper
 transcriptor = Transcriptor(file, log_folder, language, modelSize, ACCESS_TOKEN, voices_folder, quantization)
@@ -24,4 +25,4 @@
 res = transcriptor.huggingface_model("Jingmiao/whisper-small-chinese_base")
 
 # use assembly ai model
-res = transcriptor.assemby_ai_model("your api key")
+res = transcriptor.assemby_ai_model("assemblyAI api key")

library.md

@@ -70,13 +70,16 @@ transcript will also indicate the timeframe in seconds where each speaker speaks
 ### Transcription example:
 
 ```
+import os
+from speechlib import Transcriptor
+
 file = "obama_zach.wav"  # your audio file
 voices_folder = "" # voices folder containing voice samples for recognition
 language = "en"          # language code
 log_folder = "logs"      # log folder for storing transcripts
 modelSize = "tiny"     # size of model to be used [tiny, small, medium, large-v1, large-v2, large-v3]
 quantization = False   # setting this 'True' may speed up the process but lower the accuracy
-ACCESS_TOKEN = "your hf key" # get permission to access pyannote/[email protected] on huggingface
+ACCESS_TOKEN = "huggingface api key" # get permission to access pyannote/[email protected] on huggingface
 
 # quantization only works on faster-whisper
 transcriptor = Transcriptor(file, log_folder, language, modelSize, ACCESS_TOKEN, voices_folder, quantization)
@@ -94,7 +97,9 @@ res = transcriptor.custom_whisper("D:/whisper_tiny_model/tiny.pt")
 res = transcriptor.huggingface_model("Jingmiao/whisper-small-chinese_base")
 
 # use assembly ai model
-res = transcriptor.assemby_ai_model("your api key")
+res = transcriptor.assemby_ai_model("assemblyAI api key")
+
+res --> [["start", "end", "text", "speaker"], ["start", "end", "text", "speaker"]...]
 ```
 
 #### if you don't want speaker names: keep voices_folder as an empty string ""

requirements.txt

@@ -1,9 +1,9 @@
-transformers==4.36.2
-torch==2.1.2
-torchaudio==2.1.2
-pydub==0.25.1
-pyannote.audio==3.1.1
-speechbrain==0.5.16
-accelerate==0.26.1
-faster-whisper==0.10.1
-openai-whisper==20231117
+transformers>=4.36.2, <5.0.0
+torch>=2.1.2, <3.0.0
+torchaudio>=2.1.2, <3.0.0
+pydub>=0.25.1, <1.0.0
+pyannote.audio>=3.1.1, <4.0.0
+speechbrain>=0.5.16, <1.0.0
+accelerate>=0.26.1, <1.0.0
+faster-whisper>=0.10.1, <1.0.0
+openai-whisper>=20231117, <20240927

setup.py

@@ -5,7 +5,7 @@
 
 setup(
     name="speechlib",
-    version="1.1.9",  
+    version="1.1.10",  
     description="speechlib is a library that can do speaker diarization, transcription and speaker recognition on an audio file to create transcripts with actual speaker names. This library also contain audio preprocessor functions.",
     packages=find_packages(),
     long_description=long_description,
@@ -19,7 +19,7 @@
         "Programming Language :: Python :: 3.10",
         "Operating System :: OS Independent",
     ],
-    install_requires=["transformers", "torch", "torchaudio", "pydub", "pyannote.audio", "speechbrain==0.5.16", "accelerate", "faster-whisper", "openai-whisper", "assemblyai"],
+    install_requires=["transformers>=4.36.2, <5.0.0", "torch>=2.1.2, <3.0.0", "torchaudio>=2.1.2, <3.0.0", "pydub>=0.25.1, <1.0.0", "pyannote.audio>=3.1.1, <4.0.0", "speechbrain>=0.5.16, <1.0.0", "accelerate>=0.26.1, <1.0.0", "faster-whisper>=0.10.1, <1.0.0", "openai-whisper>=20231117, <20240927", "assemblyai"],
     python_requires=">=3.8",
 )
 

setup_instruction.md

@@ -9,7 +9,7 @@ for publishing:
     pip install twine
 
 for install locally for testing:
-    pip install dist/speechlib-1.1.9-py3-none-any.whl
+    pip install dist/speechlib-1.1.10-py3-none-any.whl
 
 finally run:
     twine upload dist/*