scribe-cli 0.4.1__tar.gz → 0.5.1__tar.gz

{scribe_cli-0.4.1/scribe_cli.egg-info → scribe_cli-0.5.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: scribe-cli
-Version: 0.4.1
+Version: 0.5.1
 Summary: scribe is a local speech recognition tool that provides real-time transcription using vosk and whisper AI.
 Author-email: Mahé Perrette <mahe.perrette@gmail.com>
 License: MIT License
@@ -43,6 +43,7 @@ Requires-Dist: numpy
 Requires-Dist: sounddevice
 Requires-Dist: tqdm
 Requires-Dist: requests
+Requires-Dist: pyperclip
 Provides-Extra: keyboard
 Requires-Dist: pynput; extra == "keyboard"
 Provides-Extra: whisper
@@ -56,7 +57,7 @@ Requires-Dist: vosk; extra == "all"
 
 # Scribe
 
-`scribe` is a local speech recognition tool that provides real-time transcription using vosk and whisper AI.
+`scribe` is a local speech recognition tool that provides real-time transcription using vosk and whisper AI, with the goal of serving as a virtual keyboard.
 
 ## Installation
 
@@ -99,7 +100,7 @@ scribe
 and the script will guide you through the choice of backend (`whisper` or `vosk`) and the specific language model.
 After this, you will be prompted to start recording your microphone and print the transcribed text in real-time (`vosk`)
 or until after recording is complete (`whisper`).
-You can interrupt the recording via Ctrl + C and start again or change model.
+You can interrupt the recording via Ctrl + C and start again or change model. The full content of the transcription will be pasted to the clipboard by default, until interruption.
 
 The default (`whisper`) is excellent at transcribing a full-length audio sequences in [many languages](https://github.com/openai/whisper?tab=readme-ov-file#available-models-and-languages). It is really impressive,
 but it cannot do real-time, and depending on the model can have relatively long execution time, especially with the `turbo` model (at least on my laptop with CPU only). The `small` model is also excellent and runs much faster. It is selected as default in `scribe` for that reason.
@@ -118,6 +119,7 @@ where `--no-prompt` jumps right to the recording (after the first interruption,
 
 ### Advanced usage as keyboard replacement
 
+By default the content of the transcription is paster to the clipboard, but is not propagated further.
 With the `--keyboard` option `scribe` will attempt to simulate a keyboard and send transcribed characters to the applcation under focus:
 
 ```bash

{scribe_cli-0.4.1 → scribe_cli-0.5.1}/README.md

@@ -1,6 +1,6 @@
 # Scribe
 
-`scribe` is a local speech recognition tool that provides real-time transcription using vosk and whisper AI.
+`scribe` is a local speech recognition tool that provides real-time transcription using vosk and whisper AI, with the goal of serving as a virtual keyboard.
 
 ## Installation
 
@@ -43,7 +43,7 @@ scribe
 and the script will guide you through the choice of backend (`whisper` or `vosk`) and the specific language model.
 After this, you will be prompted to start recording your microphone and print the transcribed text in real-time (`vosk`)
 or until after recording is complete (`whisper`).
-You can interrupt the recording via Ctrl + C and start again or change model.
+You can interrupt the recording via Ctrl + C and start again or change model. The full content of the transcription will be pasted to the clipboard by default, until interruption.
 
 The default (`whisper`) is excellent at transcribing a full-length audio sequences in [many languages](https://github.com/openai/whisper?tab=readme-ov-file#available-models-and-languages). It is really impressive,
 but it cannot do real-time, and depending on the model can have relatively long execution time, especially with the `turbo` model (at least on my laptop with CPU only). The `small` model is also excellent and runs much faster. It is selected as default in `scribe` for that reason.
@@ -62,6 +62,7 @@ where `--no-prompt` jumps right to the recording (after the first interruption,
 
 ### Advanced usage as keyboard replacement
 
+By default the content of the transcription is paster to the clipboard, but is not propagated further.
 With the `--keyboard` option `scribe` will attempt to simulate a keyboard and send transcribed characters to the applcation under focus:
 
 ```bash

{scribe_cli-0.4.1 → scribe_cli-0.5.1}/pyproject.toml

@@ -17,6 +17,7 @@ dependencies = [
     "sounddevice",
     "tqdm",
     "requests",
+    "pyperclip",
 ]
 optional-dependencies = { keyboard = ["pynput"], whisper = ["openai-whisper"], vosk = ["vosk"], all = ["pynput", "openai-whisper", "vosk"] }
 

{scribe_cli-0.4.1 → scribe_cli-0.5.1}/scribe/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.4.1'
-__version_tuple__ = version_tuple = (0, 4, 1)
+__version__ = version = '0.5.1'
+__version_tuple__ = version_tuple = (0, 5, 1)

{scribe_cli-0.4.1 → scribe_cli-0.5.1}/scribe/models.py

@@ -1,6 +1,7 @@
 import os
 import json
 import numpy as np
+import time
 from scribe.util import download_model
 
 VOSK_MODELS_FOLDER = os.path.join(os.environ.get("HOME"),
@@ -9,21 +10,22 @@ VOSK_MODELS_FOLDER = os.path.join(os.environ.get("HOME"),
 
 class AbstractTranscriber:
     backend = None
-    def __init__(self, model, model_name=None, language=None, samplerate=16000, max_duration=None, model_kwargs={}):
+    def __init__(self, model, model_name=None, language=None, samplerate=16000, timeout=None, model_kwargs={}):
         self.model_name = model_name
         self.language = language
         self.model = model
         self.model_kwargs = model_kwargs
         self.samplerate = samplerate
-        self.max_duration = max_duration
+        self.timeout = timeout
         self.one_second_bytes = self.samplerate * 2 # 16-bit audio, 1 channel  ~ 32000 bytes
         self.audio_buffer = b''
+        self.start_time = time.time()
 
-    def get_elapsed(self, size=None):
-        return len(size or self.audio_buffer) / self.one_second_bytes
+    def get_elapsed(self):
+        return time.time() - self.start_time
 
-    def is_overtime(self, elapsed=None, size=None):
-        return self.max_duration and (elapsed or self.get_elapsed(size)) > self.max_duration
+    def is_overtime(self):
+        return time.time() - self.start_time > self.timeout
 
     def transcribe_realtime_audio(self, audio_bytes=b""):
         self.audio_buffer += audio_bytes

{scribe_cli-0.4.1 → scribe_cli-0.5.1}/scribe/streamer.py

@@ -12,14 +12,25 @@ language_config = language_config_default.copy()
 
 
 # Commencer l'enregistrement
-def start_recording(micro, transcriber, keyboard=False, latency=0):
+def start_recording(micro, transcriber, clipboard=True, keyboard=False, latency=0):
 
     if keyboard:
         try:
             from scribe.keyboard import type_text
         except ImportError:
             keyboard = False
-            exit(1)
+            print("Keyboard simulation is not available.")
+            return
+
+    if clipboard:
+        try:
+            import pyperclip
+        except ImportError:
+            clipboard = False
+            print("Clipboard simulation is not available.")
+            return
+
+    fulltext = ""
 
     greetings = { k: v for k, v in language_config["_meta"].get(transcriber.language, {}).items()
                 if v is not None and k.startswith(("start", "stop"))
@@ -32,6 +43,11 @@ def start_recording(micro, transcriber, keyboard=False, latency=0):
             print(result.get('text'))
             if keyboard:
                 type_text(result['text'] + " ", interval=latency) # Simulate typing
+
+            if clipboard:
+                fulltext += result['text'] + " "
+                pyperclip.copy(fulltext)
+
         else:
             print_partial(result.get('partial', ''))
 
@@ -136,7 +152,7 @@ def get_transcriber(o, prompt=True):
             transcriber = VoskTranscriber(model_name=model,
                                         language=o.language,
                                         samplerate=o.samplerate,
-                                        max_duration=None, # vosk keeps going (no timeout)
+                                        timeout=None, # vosk keeps going (no timeout)
                                         model_kwargs={"data_folder": o.data_folder})
         except Exception as error:
             print(error)
@@ -144,7 +160,7 @@ def get_transcriber(o, prompt=True):
             exit(1)
 
     elif backend == "whisper":
-        transcriber = WhisperTranscriber(model_name=model, language=o.language, samplerate=o.samplerate, max_duration=o.duration)
+        transcriber = WhisperTranscriber(model_name=model, language=o.language, samplerate=o.samplerate, timeout=o.duration)
 
     else:
         raise ValueError(f"Unknown backend: {backend}")
@@ -170,6 +186,7 @@ def get_parser():
     parser.add_argument("--samplerate", default=16000, type=int, help=argparse.SUPPRESS)
     parser.add_argument("--duration", default=60, type=int, help="duration in seconds before whisper models start transcribing (default %(default)ss)")
     parser.add_argument("--keyboard", action="store_true")
+    parser.add_argument("--no-clipboard", dest="clipboard", action="store_false")
     parser.add_argument("--latency", default=0, type=float, help="keyboard latency")
 
     parser.add_argument("--data-folder", help="Folder to store Vosk models.")
@@ -191,14 +208,15 @@ def main(args=None):
     while True:
         if transcriber is None:
             transcriber = get_transcriber(o, prompt=o.prompt)
-        print(f"[ Model {transcriber.model_name} from {transcriber.backend} selected. ]")
+        print(f"[ Model {transcriber.model_name} from {transcriber.backend} selected. Keyboard [{'on' if o.keyboard else 'off'}]. Clipboard [{'on' if o.clipboard else 'off'}]]")
         if o.prompt:
             print(f"Choose any of the following actions:")
             print(f"[q] quit")
             print(f"[e] change model")
-            print(f"[k] toggle keyboard {'off' if o.keyboard else 'on'}")
+            print(f"[k] toggle keyboard [{'off' if o.keyboard else 'on'}]")
+            print(f"[c] toggle clipboard [{'off' if o.clipboard else 'on'}]")
             if transcriber.backend == "whisper":
-                print(f"[t] change duration (currently {transcriber.max_duration}s)")
+                print(f"[t] change duration (currently {transcriber.timeout}s)")
             print(colored(f"Press [Enter] or any other key to start recording.", "BOLD"))
 
             key = input()
@@ -210,15 +228,18 @@ def main(args=None):
             if key == "k":
                 o.keyboard = not o.keyboard
                 continue
+            if key == "c":
+                o.clipboard = not o.clipboard
+                continue
             if key == "t":
-                duration = input(f"Enter new duration in seconds (current: {transcriber.max_duration}): ")
+                duration = input(f"Enter new duration in seconds (current: {transcriber.timeout}): ")
                 try:
-                    o.duration = transcriber.max_duration = int(duration)
+                    o.duration = transcriber.timeout = int(duration)
                 except:
                     print("Invalid duration. Must be an integer.")
                 continue
 
-        start_recording(micro, transcriber, keyboard=o.keyboard, latency=o.latency)
+        start_recording(micro, transcriber, clipboard=o.clipboard, keyboard=o.keyboard, latency=o.latency)
 
         # if we arrived so far, that means we pressed Ctrl + C anyway, and need Enter to move on.
         # So we leave the wider range of options to change the model.

{scribe_cli-0.4.1 → scribe_cli-0.5.1/scribe_cli.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: scribe-cli
-Version: 0.4.1
+Version: 0.5.1
 Summary: scribe is a local speech recognition tool that provides real-time transcription using vosk and whisper AI.
 Author-email: Mahé Perrette <mahe.perrette@gmail.com>
 License: MIT License
@@ -43,6 +43,7 @@ Requires-Dist: numpy
 Requires-Dist: sounddevice
 Requires-Dist: tqdm
 Requires-Dist: requests
+Requires-Dist: pyperclip
 Provides-Extra: keyboard
 Requires-Dist: pynput; extra == "keyboard"
 Provides-Extra: whisper
@@ -56,7 +57,7 @@ Requires-Dist: vosk; extra == "all"
 
 # Scribe
 
-`scribe` is a local speech recognition tool that provides real-time transcription using vosk and whisper AI.
+`scribe` is a local speech recognition tool that provides real-time transcription using vosk and whisper AI, with the goal of serving as a virtual keyboard.
 
 ## Installation
 
@@ -99,7 +100,7 @@ scribe
 and the script will guide you through the choice of backend (`whisper` or `vosk`) and the specific language model.
 After this, you will be prompted to start recording your microphone and print the transcribed text in real-time (`vosk`)
 or until after recording is complete (`whisper`).
-You can interrupt the recording via Ctrl + C and start again or change model.
+You can interrupt the recording via Ctrl + C and start again or change model. The full content of the transcription will be pasted to the clipboard by default, until interruption.
 
 The default (`whisper`) is excellent at transcribing a full-length audio sequences in [many languages](https://github.com/openai/whisper?tab=readme-ov-file#available-models-and-languages). It is really impressive,
 but it cannot do real-time, and depending on the model can have relatively long execution time, especially with the `turbo` model (at least on my laptop with CPU only). The `small` model is also excellent and runs much faster. It is selected as default in `scribe` for that reason.
@@ -118,6 +119,7 @@ where `--no-prompt` jumps right to the recording (after the first interruption,
 
 ### Advanced usage as keyboard replacement
 
+By default the content of the transcription is paster to the clipboard, but is not propagated further.
 With the `--keyboard` option `scribe` will attempt to simulate a keyboard and send transcribed characters to the applcation under focus:
 
 ```bash

{scribe_cli-0.4.1 → scribe_cli-0.5.1}/scribe_cli.egg-info/requires.txt

@@ -2,6 +2,7 @@ numpy
 sounddevice
 tqdm
 requests
+pyperclip
 
 [all]
 pynput