scribe-cli 0.6.0__tar.gz → 0.6.2__tar.gz

{scribe_cli-0.6.0/scribe_cli.egg-info → scribe_cli-0.6.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: scribe-cli
-Version: 0.6.0
+Version: 0.6.2
 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
@@ -109,6 +109,7 @@ there is a maximum duration after which it will stop by itself, which is setup t
 
 The `vosk` backend is good at
 doing real-time transcription for one language, but tended to make more mistakes in my tests and it does not do punctuation.
+Use mainly for longer typing session with the [keyboard](#virtual-keyboard-advanced) option, e.g. to make notes.
 There are many [vosk models](https://alphacephei.com/vosk/models) available, and here a few are associated to [a handful of languages](scribe/models.toml) `en`, `fr`, `it`, `de` (so far).
 
 To skip the initial selection menu you can do:
@@ -117,9 +118,9 @@ scribe --backend whisper --model small --no-prompt
 ```
 where `--no-prompt` jumps right to the recording (after the first interruption, you can still choose to change the backend and model).
 
-### Advanced usage as keyboard replacement
+### Virtual keyboard (experimental)
 
-By default the content of the transcription is paster to the clipboard, but is not propagated further.
+By default the content of the transcription is pasted to the clipboard, and it is up to the user to paste (e.g. Ctrl + V).
 With the `--keyboard` option `scribe` will attempt to simulate a keyboard and send transcribed characters to the applcation under focus:
 
 ```bash
@@ -128,10 +129,18 @@ scribe --keyboard
 
 It relies on the optional `pynput` dependency (installed together with `scribe` if you used the `[all]` or `[keyboard]` option).
 
-`pynput` may require [some configuration](https://pynput.readthedocs.io/en/latest/limitations.html) (I *think* got it to work with `xhost +SI:localuser:$(whoami)` as far as the display is concerned). It has [limitations]((https://pynput.readthedocs.io/en/latest/limitations.html)). In my Ubuntu + Wayland system it works in chromium based applications (including vscode) but it does not in firefox and sublime text and any of the rest (not even in a terminal !).
-Workarounds include using the Xorg version of GNOME... Suggestions welcome.
+`pynput` may require [some configuration](https://pynput.readthedocs.io/en/latest/limitations.html). It has [limitations]((https://pynput.readthedocs.io/en/latest/limitations.html)). In my Ubuntu + Wayland system it works in chromium based applications (including vscode) but it does not in firefox and sublime text and any of the rest (not even in a terminal !). Workarounds include using the Xorg version of GNOME: in `etc/gdm3/custom.conf` uncomment `# WaylandEnable=false` and restart.
+
 
 ### Start as an application in Ubuntu
 
 If you run Ubuntu (or else?) with GNOME, the script `scribe-install [...]` will create a `scribe.desktop` file and place it under `$HOME/.local/share/applications`
 to make it available from the quick launch menu. Any option will be passed on to `scribe`.
+
+e.g.
+
+```bash
+scribe-install --backend whisper --model small
+```
+
+After that just typing Cmd + scri... at any time from any where will conveniently start the app in its own terminal with the prescribed options.

{scribe_cli-0.6.0 → scribe_cli-0.6.2}/README.md

@@ -52,6 +52,7 @@ there is a maximum duration after which it will stop by itself, which is setup t
 
 The `vosk` backend is good at
 doing real-time transcription for one language, but tended to make more mistakes in my tests and it does not do punctuation.
+Use mainly for longer typing session with the [keyboard](#virtual-keyboard-advanced) option, e.g. to make notes.
 There are many [vosk models](https://alphacephei.com/vosk/models) available, and here a few are associated to [a handful of languages](scribe/models.toml) `en`, `fr`, `it`, `de` (so far).
 
 To skip the initial selection menu you can do:
@@ -60,9 +61,9 @@ scribe --backend whisper --model small --no-prompt
 ```
 where `--no-prompt` jumps right to the recording (after the first interruption, you can still choose to change the backend and model).
 
-### Advanced usage as keyboard replacement
+### Virtual keyboard (experimental)
 
-By default the content of the transcription is paster to the clipboard, but is not propagated further.
+By default the content of the transcription is pasted to the clipboard, and it is up to the user to paste (e.g. Ctrl + V).
 With the `--keyboard` option `scribe` will attempt to simulate a keyboard and send transcribed characters to the applcation under focus:
 
 ```bash
@@ -71,10 +72,18 @@ scribe --keyboard
 
 It relies on the optional `pynput` dependency (installed together with `scribe` if you used the `[all]` or `[keyboard]` option).
 
-`pynput` may require [some configuration](https://pynput.readthedocs.io/en/latest/limitations.html) (I *think* got it to work with `xhost +SI:localuser:$(whoami)` as far as the display is concerned). It has [limitations]((https://pynput.readthedocs.io/en/latest/limitations.html)). In my Ubuntu + Wayland system it works in chromium based applications (including vscode) but it does not in firefox and sublime text and any of the rest (not even in a terminal !).
-Workarounds include using the Xorg version of GNOME... Suggestions welcome.
+`pynput` may require [some configuration](https://pynput.readthedocs.io/en/latest/limitations.html). It has [limitations]((https://pynput.readthedocs.io/en/latest/limitations.html)). In my Ubuntu + Wayland system it works in chromium based applications (including vscode) but it does not in firefox and sublime text and any of the rest (not even in a terminal !). Workarounds include using the Xorg version of GNOME: in `etc/gdm3/custom.conf` uncomment `# WaylandEnable=false` and restart.
+
 
 ### Start as an application in Ubuntu
 
 If you run Ubuntu (or else?) with GNOME, the script `scribe-install [...]` will create a `scribe.desktop` file and place it under `$HOME/.local/share/applications`
-to make it available from the quick launch menu. Any option will be passed on to `scribe`.
+to make it available from the quick launch menu. Any option will be passed on to `scribe`.
+
+e.g.
+
+```bash
+scribe-install --backend whisper --model small
+```
+
+After that just typing Cmd + scri... at any time from any where will conveniently start the app in its own terminal with the prescribed options.

{scribe_cli-0.6.0 → scribe_cli-0.6.2}/scribe/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.6.0'
-__version_tuple__ = version_tuple = (0, 6, 0)
+__version__ = version = '0.6.2'
+__version_tuple__ = version_tuple = (0, 6, 2)

{scribe_cli-0.6.0 → scribe_cli-0.6.2}/scribe/keyboard.py

@@ -6,7 +6,6 @@ try:
 
 except ImportError:
     print("Please install pynput to use the keyboard feature.")
-    print("Alternatively specify [keyboard] optional dependency to voskrealtime, e.g. `pip install -e .[keyboard]`")
     raise
 
 # Create a keyboard controller

{scribe_cli-0.6.0 → scribe_cli-0.6.2}/scribe/models.py

@@ -95,16 +95,16 @@ class AbstractTranscriber:
             print(stop_message)
 
 
-def get_vosk_model(model, data_folder=None, url=None):
+def get_vosk_model(model, download_root=None, url=None):
     """Load the Vosk recognizer"""
     import vosk
-    if data_folder is None:
-        data_folder = VOSK_MODELS_FOLDER
-    model_path = os.path.join(data_folder, model)
+    if download_root is None:
+        download_root = VOSK_MODELS_FOLDER
+    model_path = os.path.join(download_root, model)
     if not os.path.exists(model_path):
         if url is None:
             url = f"https://alphacephei.com/vosk/models/{model}.zip"
-        download_model(url, data_folder)
+        download_model(url, download_root)
         assert os.path.exists(model_path)
 
     return vosk.Model(model_path)
@@ -162,11 +162,11 @@ class WhisperTranscriber(AbstractTranscriber):
     def __init__(self, model_name, language=None, model=None, model_kwargs={}, **kwargs):
         import whisper
         if model is None:
-            model = whisper.load_model(model_name)
+            model = whisper.load_model(model_name, **model_kwargs)
         super().__init__(model, model_name, language, model_kwargs=model_kwargs, **kwargs)
 
     def transcribe_audio(self, audio_bytes):
-        print("Transcribing...")
+        print("\nTranscribing...")
         audio_array = np.frombuffer(audio_bytes, dtype=np.int16).flatten().astype(np.float32) / 32768.0
         return self.model.transcribe(audio_array, fp16=False, language=self.language)
 

{scribe_cli-0.6.0 → scribe_cli-0.6.2}/scribe/streamer.py

@@ -47,7 +47,7 @@ def get_transcriber(o, prompt=True):
         backend = o.backend
 
     elif not prompt:
-        backend = choices[0]
+        backend = BACKENDS[0]
 
     else:
         checked_backend = False
@@ -113,14 +113,16 @@ def get_transcriber(o, prompt=True):
                                         samplerate=o.samplerate,
                                         timeout=None, # vosk keeps going (no timeout)
                                         silence_duration=None, # vosk handles silences internally
-                                        model_kwargs={"data_folder": o.data_folder})
+                                        model_kwargs={"download_root": o.download_folder_vosk})
         except Exception as error:
             print(error)
             print(f"Failed to (down)load model {model}.")
             exit(1)
 
     elif backend == "whisper":
-        transcriber = WhisperTranscriber(model_name=model, language=o.language, samplerate=o.samplerate, timeout=o.duration, silence_duration=o.silence)
+        transcriber = WhisperTranscriber(model_name=model, language=o.language, samplerate=o.samplerate,
+                                         timeout=o.duration, silence_duration=o.silence, restart_after_silence=o.restart_after_silence,
+                                         model_kwargs={"download_root": o.download_folder_whisper})
 
     else:
         raise ValueError(f"Unknown backend: {backend}")
@@ -153,7 +155,8 @@ def get_parser():
     group.add_argument("--silence", default=2, type=float, help="silence duration that prompt transcription (whisper) (default %(default)ss)")
     group.add_argument("--restart-after-silence", action="store_true", help="Restart the recording after a transcription triggered by a silence")
 
-    parser.add_argument("--data-folder", help="Folder to store Vosk models.")
+    parser.add_argument("--download-folder-vosk", help="Folder to store Vosk models.")
+    parser.add_argument("--download-folder-whisper", help="Folder to store Whisper models.")
 
     return parser
 
@@ -235,7 +238,7 @@ def main(args=None):
             if transcriber.backend == "whisper":
                 print(f"[t] change duration (currently {transcriber.timeout}s)")
                 print(f"[b] change silence duration (currently {transcriber.silence_duration}s)")
-                print(f"[a] toggle auto-restart after silence [{toggle[o.restart_after_silence]}] -> [{toggle[not o.restart_after_silence]}]")
+                print(f"[a] toggle auto-restart after silence [{toggle[transcriber.restart_after_silence]}] -> [{toggle[not transcriber.restart_after_silence]}]")
             print(colored(f"Press [Enter] or any other key to start recording.", "BOLD"))
 
             key = input()
@@ -250,6 +253,9 @@ def main(args=None):
             if key == "c":
                 o.clipboard = not o.clipboard
                 continue
+            if key == "a":
+                transcriber.restart_after_silence = not transcriber.restart_after_silence
+                continue
             if key == "t":
                 ans = input(f"Enter new duration in seconds (current: {transcriber.timeout}): ")
                 try:

{scribe_cli-0.6.0 → scribe_cli-0.6.2/scribe_cli.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: scribe-cli
-Version: 0.6.0
+Version: 0.6.2
 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
@@ -109,6 +109,7 @@ there is a maximum duration after which it will stop by itself, which is setup t
 
 The `vosk` backend is good at
 doing real-time transcription for one language, but tended to make more mistakes in my tests and it does not do punctuation.
+Use mainly for longer typing session with the [keyboard](#virtual-keyboard-advanced) option, e.g. to make notes.
 There are many [vosk models](https://alphacephei.com/vosk/models) available, and here a few are associated to [a handful of languages](scribe/models.toml) `en`, `fr`, `it`, `de` (so far).
 
 To skip the initial selection menu you can do:
@@ -117,9 +118,9 @@ scribe --backend whisper --model small --no-prompt
 ```
 where `--no-prompt` jumps right to the recording (after the first interruption, you can still choose to change the backend and model).
 
-### Advanced usage as keyboard replacement
+### Virtual keyboard (experimental)
 
-By default the content of the transcription is paster to the clipboard, but is not propagated further.
+By default the content of the transcription is pasted to the clipboard, and it is up to the user to paste (e.g. Ctrl + V).
 With the `--keyboard` option `scribe` will attempt to simulate a keyboard and send transcribed characters to the applcation under focus:
 
 ```bash
@@ -128,10 +129,18 @@ scribe --keyboard
 
 It relies on the optional `pynput` dependency (installed together with `scribe` if you used the `[all]` or `[keyboard]` option).
 
-`pynput` may require [some configuration](https://pynput.readthedocs.io/en/latest/limitations.html) (I *think* got it to work with `xhost +SI:localuser:$(whoami)` as far as the display is concerned). It has [limitations]((https://pynput.readthedocs.io/en/latest/limitations.html)). In my Ubuntu + Wayland system it works in chromium based applications (including vscode) but it does not in firefox and sublime text and any of the rest (not even in a terminal !).
-Workarounds include using the Xorg version of GNOME... Suggestions welcome.
+`pynput` may require [some configuration](https://pynput.readthedocs.io/en/latest/limitations.html). It has [limitations]((https://pynput.readthedocs.io/en/latest/limitations.html)). In my Ubuntu + Wayland system it works in chromium based applications (including vscode) but it does not in firefox and sublime text and any of the rest (not even in a terminal !). Workarounds include using the Xorg version of GNOME: in `etc/gdm3/custom.conf` uncomment `# WaylandEnable=false` and restart.
+
 
 ### Start as an application in Ubuntu
 
 If you run Ubuntu (or else?) with GNOME, the script `scribe-install [...]` will create a `scribe.desktop` file and place it under `$HOME/.local/share/applications`
 to make it available from the quick launch menu. Any option will be passed on to `scribe`.
+
+e.g.
+
+```bash
+scribe-install --backend whisper --model small
+```
+
+After that just typing Cmd + scri... at any time from any where will conveniently start the app in its own terminal with the prescribed options.