SpeechPulse Updates

7/16/2024 - Version 5.2.0 (Windows)

  • History file list now supports navigation with arrow keys.
  • A new button to open the history folder in file explorer.
  • Re-added custom prompts support.

7/14/2024 - Version 5.1.0 (macOS)

  • Notify the user when new SpeechPulse versions are available.

7/13/2024 - Version 5.1.0 (Windows)

  • Notify the user when new SpeechPulse versions are available.
  • Fixed an issue that caused SpeechPulse to reset model settings.

7/12/2024 - Version 5.0.0 (macOS)

  • A new UI with a modern look.
  • Misc. improvements.

7/10/2024 - Version 5.0.0 (Windows)

  • A new UI with a modern look.
  • Misc. improvements.

Live Mode New live mode UI File Mode New file mode UI History View New history mode UI Speaker Diarization View New Speaker Diarization UI

7/2/2024 - Version 4.5.6 (Windows)

  • Fixed an issue that caused SpeechPulse to crash silently in some specific CPUs and Windows versions.

6/27/2024 - Version 4.5.3 (Windows and macOS)

  • SpeechPulse can now process the currently copied text in the clipboard using AI language models. First, use CTRL+C to copy the text to the clipboard. Then, start clipboard processing using the clipboard processing hotkey (configurable via "Settings->Options->Hotkeys").
  • Fixed an issue that caused incorrect formatting of phone numbers.

6/25/2024 - Version 4.5.2 (Windows)

  • Fixed an issue that caused UI artifacts when moving SpeechPulse UI between two monitors with different DPI scaling.

6/22/2024 - Version 4.5.1 (Windows and macOS)

  • Fixed an issue that caused significant text hallucinations and poor accuracy in the Auto punctuation mode.
  • The automatic microphone on/off feature in the push-to-talk mode is now optional. It can be enabled/disabled via "Settings->Options->General settings".

6/20/2024 - Version 4.5.0 (Windows)

  • SpeechPulse (Windows) now comes with a built-in model/library downloader.
  • Model downloader can be used to download speech models, language models, and CUDA GPU libraries.

6/16/2024 - Version 4.4.0 (macOS)

  • SpeechPulse (macOS) now supports OpenAI compatible Whisper speech APIs.
  • Fixed an issue that prevented AI template editing.
  • UI improvements.

6/14/2024 - Version 4.4.0 (Windows)

  • SpeechPulse (Windows) now supports OpenAI compatible Whisper speech APIs.
  • UI improvements.

6/11/2024 - Version 4.3.1 (macOS)

  • SpeechPulse (macOS) now supports OpenAI compatible external AI language (LLM) APIs. For example, you can connect to your Ollama server using SpeechPulse.
  • Improved internal implementation for the AI language (LLM) feature. Now delivers more precise results for your AI templates.
  • Push-to-talk mode now automatically turns on/off the microphone.
  • A new option to keep the speech model loaded in RAM for faster transcription.
  • Misc. improvements.

6/09/2024 - Version 4.3.1 (Windows)

  • Push-to-talk mode now automatically turns on/off the microphone.
  • Better error handling when connecting to LLM APIs.
  • Fixed an issue that caused the auto-microphone-off feature to stop working.
  • UI improvements.

6/08/2024 - Version 4.3.0 (Windows)

  • SpeechPulse (Windows) now supports OpenAI compatible external AI language (LLM) APIs. For example, you can connect to your Ollama server using SpeechPulse.
  • Improved internal implementation for the AI language (LLM) feature. Now delivers more precise results for your AI templates.
  • Fixed an issue that caused missing words near the diarization boundaries.
  • Misc. improvements.

6/03/2024 - Version 4.2.0 (macOS)

  • SpeechPulse (macOS) now supports automatic speaker diarization.
  • File mode now supports generating all output formats in a single pass.
  • A new option to limit the number of words per subtitle line.
  • UI improvements.

5/26/2024 - Version 4.1.4 (macOS)

  • You can configure how many days of previous recordings to keep or disable history recordings via "Settings->Options->General settings". (this option was missing in the previous version)
  • A new indicator to inform users when the push-to-talk mode is active.

5/26/2024 - Version 4.2.0 (Windows)

  • SpeechPulse (Windows) now supports automatic speaker diarization.
  • Supports both CPU and GPU execution for diarization.
  • File mode now supports generating all output formats in a single pass.
  • A new option to limit the number of words per subtitle line.
  • A new indicator to inform users when the push-to-talk mode is active.
  • UI improvements.
  • Removed the support for onnx models due to their limitations.
SpeechPulse speaker diarization

5/16/2024 - Version 4.1.3 (macOS)

  • SpeechPulse now supports recording your dictations as WAV files inside the "Documents/SpeechPulse" folder.
  • You can configure how many days of previous recordings to keep or disable history recordings via "Settings->Options->General settings".
  • The new history window allows you to process your previous dictation recordings using different speech models, language models, AI Templates, or any other different settings.
  • Now supports importing other compatible offline language models. For example, you can run LLAMA 3 using SpeechPulse. Just place the GGUF language model file inside the SpeechPulse models directory. SpeechPulse will automatically detect the model on startup. For AI language models on macOS, it's recommended to have at least 16GB of RAM.
  • Fixed an issue that caused SpeechPulse to skip AI formatting when dictated to the built-in editor.
  • UI changes for better usability on high DPI screens.

5/14/2024 - Version 4.0.1 (macOS)

  • Fixed an issue that caused SpeechPulse to output text in all lowercase letters in the manual punctuation mode.

5/14/2024 - Version 4.1.3 (Windows)

  • Fixed an issue that caused SpeechPulse to output text in all lowercase letters in the manual punctuation mode.

5/12/2024 - Version 4.1.2 (Windows)

  • Fixed an issue that caused SpeechPulse to crash with the latest NVIDIA drivers. Now supports all the recent driver versions.
  • UI improvements.

5/10/2024 - Version 4.1.0 (Windows)

  • SpeechPulse now supports recording your dictations as WAV files inside the "Documents/SpeechPulse" folder.
  • You can configure how many days of previous recordings to keep or disable history recordings via "Settings->Options->General settings".
  • The new history window allows you to process your previous dictation recordings using different speech models, language models, AI Templates, or any other different settings.
  • Now supports importing other compatible offline language models. For example, you can run LLAMA 3 using SpeechPulse. Just place the GGUF language model file inside the SpeechPulse models directory. SpeechPulse will automatically detect the model on startup.
  • Fixed an issue that caused SpeechPulse to skip AI formatting when dictated to MS Word, WordPad, and the built-in editor.
  • UI changes for better usability on high DPI screens.
SpeechPulse dictation history window

5/05/2024 - Version 4.0.0 (macOS)

  • SpeechPulse (macOS) now supports offline AI language models for text formatting (currently only for the English language). You can use them to enhance your dictated text in real time.
  • Common use cases include grammar, spelling, and punctuation correction, summarizing text, formatting text for Email, chat, notes, etc.
  • AI language models support prompting to get the desired output. SpeechPulse has an "AI Templates" feature where you can customize the prompts for your specific use cases.
  • File mode also supports AI language models for text processing.
  • A new option to customize the deactivation delay in the Automatic speech input mode.
  • A new option to increase the context length of the AI language models (longer context lengths require more RAM).
  • Use memory more efficiently.

5/03/2024 - Version 4.0.1 (Windows)

  • A new option to increase the context length of the AI language models (longer context lengths require more RAM/VRAM).

5/02/2024

  • Replaced the original English (standard) AI language model with a slightly better one.

5/01/2024 - Version 4.0.0 (Windows)

  • SpeechPulse (Windows) now supports offline AI language models for text formatting (currently only for the English language). You can use them to enhance your dictated text in real time.
  • Common use cases include grammar, spelling, and punctuation correction, summarizing text, formatting text for Email, chat, notes, etc.
  • AI language models support prompting to get the desired output. SpeechPulse has an "AI Templates" feature where you can customize the prompts for your specific use cases.
  • AI language models support both CPU and GPU execution. However, a GPU is recommended for faster live transcription.
  • File mode also supports AI language models for text processing.
  • A new option to customize the deactivation delay in the Automatic speech input mode.

4/25/2024 - Version 3.7.2

  • Fixed an issue that caused SpeechPulse to ignore audio files with capital letters in the file extension.

4/25/2024 - Version 3.7.1

  • A new option to enable/disable automatic number formatting in the manual punctuation mode.
  • Better formatting for numbers that include the words million and billion.
  • Fixed an issue that caused incorrect date formatting.

4/24/2024 - Version 3.7.0 (macOS)

* You may have to modify your existing custom mappings with this update (especially the mappings that include the period symbol).

4/23/2024 - Version 3.7.0 (Windows)

* You may have to modify your existing custom mappings with this update (especially the mappings that include the period symbol).

  • Manual punctuation mode now supports automatically formatting numbers, dates, currency values, etc.
  • Fixed an issue that caused missing punctuation marks in the manual punctuation mode.
  • A new option to press the ENTER key after inserting text into the text edit area.
  • Allow file mode even if no microphone is connected.
  • UI improvements.

4/16/2024 - Version 3.6.9 (macOS)

  • A new option to press the ENTER key after inserting text into the text edit area.
  • Allow file mode even if no microphone is connected.
  • UI improvements.

4/14/2024 - Version 3.6.7 (macOS)

  • Fixed a crash that occurred in the activation window.
  • UI improvements.

4/10/2024 - Version 3.6.8 (Windows)

  • Fixed a crash that occurred when switching models or quitting the program on a GPU.
  • Misc. improvements.

4/9/2024 - Version 3.6.7 (Windows)

  • Fixed an issue that prevented graceful termination.
  • Misc. improvements.

4/8/2024 - Version 3.6.6 (macOS)

  • Fixed an issue that caused SpeechPulse to return a previous transcription result for failed transcriptions.
  • Preserve punctuation mode when switching languages.
  • Export/import custom mappings to easily transfer mappings from one installation to another.
  • A new controls overview window to explain different punctuation, spacing, and speech input modes.
  • Fixed an issue that caused missing words after a hyphen in the manual punctuation mode.
  • Capitalization correction now supports phrases with hyphens (e.g. Wi-Fi).
  • Fixed an issue that prevented graceful termination.
  • Misc. improvements.

4/7/2024 - Version 3.6.6 (Windows)

  • Preserve punctuation mode when switching languages.
  • Export/import custom mappings to easily transfer mappings from one installation to another.
  • A new controls overview window to explain different punctuation, spacing, and speech input modes.
  • Fixed an issue that prevented graceful termination.
  • Faster model load times.
  • Misc. improvements.

controls overview

4/5/2024 - Version 3.6.5 (Windows)

  • Fixed an issue that caused missing words after a hyphen in the manual punctuation mode.
  • Capitalization correction now supports phrases with hyphens (e.g. Wi-Fi).
  • UI improvements.

4/1/2024 - Version 3.6.4

  • Fixed an issue that caused incorrect mappings.

3/29/2024 - Version 3.6.3

  • Fixed an issue caused by incompatible character encodings on different systems.

3/29/2024 - Version 3.6.2 (macOS)

  • Fixed an issue that caused incorrect transcription for longer audio segments.
  • File mode now supports manual punctuation (For English full-text transcription).
  • Misc. improvements.

3/28/2024 - Version 3.6.2 (Windows)

  • File mode now supports manual punctuation (For English full-text transcription).
  • Faster initial load times.
  • Misc. improvements.

3/25/2024 - Version 3.6.1 (Windows)

  • On Windows, prevent hotkeys from making a notification sound.

3/24/2024 - Version 3.6.0

  • Allows you to add a list of proper names for capitalization correction.
  • Prevents subword replacements in the Mappings feature (non-RegEx mode).

3/21/2024 - Version 3.5.4 (Windows)

  • Fixed an issue that caused incorrect capitalization for the first-person pronoun "I".

3/20/2024 - Version 3.5.3

  • A new option to automatically reduce system output volume during push-to-talk.
  • Improved capitalization in the manual punctuation mode.

3/15/2024 - Version 3.5.2 (macOS)

  • Stop hotkeys and auto spacing from making a notification sound on macOS.

3/14/2024 - Version 3.5.1

  • Fixed the issue of missing period punctuation marks in the manual punctuation mode.
  • Reduce text hallucinations.

3/2/2024 - Version 3.5.0

  • Now supports manual spacing.
  • A new option to load SpeechPulse in the background and automatically minimize to the system tray.
  • A taskbar notification to indicate if the microphone is not working (need to enable via Settings).
  • Two new hotkeys for push-to-talk with auto punctuation and manual punctuation.

2/26/2024 - Version 3.4.2

  • Reduce text hallucinations.

2/21/2024 - Version 3.4.1

  • A history page to display previously dictated text of the current session.
  • Disabled Speech Profiles in the push-to-talk mode.
  • UI improvements.

2/15/2024 - Version 3.4.0

  • Push-to-talk speech input with a new hotkey.

2/12/2024 - Version 3.3.7

  • A new option to display audio buffer filling.
  • New options to enable/disable voice activity detection.
  • Now supports replacing text selections with dictation.

2/06/2024 - Version 3.3.6

  • Automatically transfer focus from the SpeechPulse window to the last active text edit.
  • Displays a "No text edit in focus!" message if there's no text edit in focus.

2/04/2024 - Version 3.3.5

  • Disabled automatic turn-off in file mode.
  • UI improvements.

2/03/2024 - Version 3.3.4

  • A new option to automatically stop listening when the user starts typing.
  • Taskbar notifications to indicate start/stop listening.
  • Notification sounds to indicate start/stop listening.
  • Now remember the file mode output format between sessions.
  • A new option to make the processing indicator white when dictating to a dark background.

2/01/2024 - Version 3.3.3

  • Now supports customizable subtitle widths in file mode.
  • Supports timestamps in live mode (With the Auto punctuation).
  • Direct text insertion in Microsoft WordPad.
  • Uses a faster implementation for type mode.

1/27/2024 - Version 3.3.2

  • Fixed an issue that prevented text generation when dictating to Microsoft Word.
  • The mappings feature now supports case-insensitive replacement and wildcards (regular expressions).
  • Better capitalization in the manual punctuation mode.

1/23/2024 - Version 3.3.1

  • If you are in a noisy environment and SpeechPulse generates random text, you can use the Speech Profiles feature to add a new speech profile that only detects your own voice.
  • Reduce text hallucinations.

1/19/2024 - Version 3.3.0

  • Now supports direct text insertion, editing, and formatting in Microsoft Word.
  • A new button to minimize SpeechPulse to the system tray.

1/14/2024 - Version 3.2.0

  • Improved capitalization in the manual punctuation mode.
  • Improved editor with voice commands for text formatting.
  • Keyboard press commands to press keys and hotkeys with your voice. (e.g. "Press Enter", "Press Control Z")
  • Displays a processing indicator and a dictated text label next to the cursor for better user experience.
  • Better and more reliable voice command implementation.
  • A new hotkey to switch punctuation modes.

12/13/2023 - Version 3.1.0

  • Now remembers language model, device, microphone, and file mode output folder settings between sessions.

12/10/2023 - Version 3.0.0

  • With the manual punctuation mode, you can dictate common punctuation marks like commas, periods, question marks, exclamation marks, colons, semicolons, etc.
  • Supports "new line" and "new paragraph" commands within a continuous speech segment.

12/02/2023 - Version 2.5.0

  • Added language hotkeys.
  • New voice command "Transfer text" for SpeechPulse editor.
  • UI improvements.

11/29/2023 - Version 2.4.0

  • Now comes with a built-in text editor.

11/14/2023 - Version 2.3.0

  • New Custom Voice Hotkeys feature can trigger custom keyboard shortcuts with voice commands.
  • Text Inserter can insert custom text snippets with voice commands.
  • Misc. improvements.

11/11/2023 - Version 2.2.0

  • New Custom Mappings feature can replace SpeechPulse's text output with your own words/phrases. For example, you can replace the phrase "speech pulse" with "SpeechPulse" using custom mappings.

11/08/2023 - Version 2.1.1

  • Trial version now supports all language models.
  • Changed the hotkey implementation.

11/06/2023 - Version 2.1.0

  • Added voice commands.
  • Lower latency in Live Mode.
  • More robust against background noise.

11/01/2023 - Version 2.0.0

  • Added hotkeys.