baristalatte/speech_to_text

テキスト整形機能を追加し、文字起こし結果を整形して返すように修正

adjust language_settings parameter

language_settings

コメント修正

細かい調整

削除時のログを一部修正

不要なログを削除

CPUまたはGPUの自動選択

コードの全体的なリファクタリング

環境変数からモデルサイズと最大ファイルサイズを取得するように変更

マイク録音時の音声プレビュー再生に関する説明を修正

ファイル削除時のエラーハンドリングを追加し、削除中の例外をキャッチしてエラーメッセージを表示

ファイル削除後に成功・失敗メッセージを表示する機能を修正

ファイル削除処理の成功・失敗メッセージを追加し、ファイルサイズ超過時のエラーメッセージを修正。ボタンクリック時の処理を簡略化。

ファイルサイズ制限を10MBから25MBに引き上げ、UIの説明を更新。音声ファイルの処理フローを改善。

一時ファイルの生成にUUIDを使用して一意性を確保し、ファイル名の衝突を防止

fix comment

マイク録音時の音声プレビュー再生に関する注意事項を追加し、利用規約への同意についての説明を明確化

Gradio UIのタイトルを更新し、利用規約に関する説明を追加。ファイルアップロードの条件を明確化。

Gradio UIのタイトルを変更し、言語選択のレイアウトを改善。ステータス表示を追加し、文字起こし結果の表示を整理。

音声ファイルの処理と文字起こし機能を改善し、エラーハンドリングを強化。Gradio UIを更新し、ユーザーに対するフィードバックを向上。

check_audioおよびtranscribe_async関数のエラーメッセージを改善し、ユーザーに対するフィードバックを強化

Gradio UIの説明文を改善し、利用規約へのリンクを明確にしました。

Gradio UIの利用規約に関する文言を更新し、サービス利用をもって利用規約に同意したとみなす旨を追記

利用規約に関する責任の明記を追加し、README.mdとterms.mdに免責事項を追記

check_audio関数のエラーハンドリングを改善し、エラー発生時に一時ファイルを削除する処理を追加

Update terms.md: Add clarifications regarding liability and service discontinuation, and include a section for users to review terms before using the service.

Update Gradio UI in app.py: Refine Japanese text for clarity, specify audio file upload limits (10MB and 10 minutes), and enhance user instructions for microphone recording and file uploads. This improves user awareness and compliance with usage guidelines.

Refactor audio validation messages and Gradio UI layout in app.py: Update Japanese text for clarity, improve file deletion success/error messages, and rearrange UI components for better user experience. Ensure consistency in language used for audio submission checks.

Enhance audio duration handling in app.py: Implement trimming for audio files exceeding duration limits, improve error messages for missing audio files, and ensure proper file deletion checks. Update Gradio UI feedback to inform users about trimming actions and submission status, enhancing overall user experience and interaction.

Refactor audio transcription process in app.py: Enhance audio validation with detailed feedback on file size and duration limits, improve error handling in transcribe_async function, and streamline Gradio UI for better user interaction. Update to use async file handling and ensure proper cleanup of temporary files, while maintaining support for both microphone recording and file uploads.

Update Gradio UI description in app.py to clarify audio file size and duration limits, enhancing user awareness and compliance with usage guidelines.

Refactor audio transcription error handling in app.py: Update transcribe_async function to return structured error messages with Gradio updates for better user feedback. Modify Gradio UI to enhance input labels and outputs, improving overall user experience and clarity in error reporting.

Refactor model loading in app.py: Update whisper model loading to use weights_only parameter for optimized performance. Modify Gradio UI to change flagging option from 'allow_flagging' to 'flagging_mode' for improved clarity in configuration.

Update Gradio UI description in app.py to include a markdown link to the terms of service for improved user awareness and compliance.

Refactor audio transcription process in app.py: Add time module for improved file deletion retries, enhance error messages for clarity, and ensure proper initialization of temporary audio file path. Update Gradio UI title and description for better user experience and compliance with terms of service.

Update transcribe_async function to support language selection (English/Japanese) with optimized settings for each language.

Refactor audio transcription process in app.py: Improve error handling and file management by using NamedTemporaryFile for secure temporary storage, enhance validation for audio file size and duration, and optimize the thread pool for Whisper model processing. Update Gradio UI for better user experience with clearer input labels and descriptions.

Update Gradio UI in app.py: Modify audio input label to improve clarity by adding a comma for better readability. This small change enhances the user interface of the audio transcription tool.

Refactor Gradio UI in app.py: Remove microphone source specification from audio input configuration to streamline user experience while maintaining support for file uploads. This change enhances clarity in the interface for the audio transcription tool.

Refactor audio transcription process in app.py: Implement asynchronous file handling with temporary storage, enhance error handling for file size and format validation, and ensure proper cleanup of temporary files. Update Gradio UI to support both microphone recording and file uploads, improving user experience and functionality.

Update app.py: Enhance description in Gradio interface to include a link to the terms of service for better user awareness and compliance.

Enhance audio format support in app.py: Expand allowed audio formats to include WAV, M4A, and OGG. Update error messages to reflect new format options for improved user guidance.

Refactor model loading in app.py: Remove weights_only parameter for clarity and update comments for better understanding of model size options. This simplifies the code and enhances maintainability.

Update app.py: Modify model loading to include weights_only parameter for improved performance, and refine Gradio UI input configuration by removing unnecessary source specification. This enhances the clarity and functionality of the audio transcription tool.

Fix typos in README, enhance disclaimer and terms of service for clarity on data handling and user responsibilities. Update model description from Tiny to Base in README. This improves user understanding of service limitations and legal obligations.

Fix model size configuration in app.py, add terms of service document, and update README with a link to the terms. This improves clarity on model usage and legal conditions for service use.

Enhance audio transcription functionality in app.py: Implement asynchronous processing, add file size and duration checks, and restrict input format to MP3. Update Gradio UI to reflect new upload constraints.

Update app.py: Enhance Whisper transcription tool UI by removing microphone source and adding title and description

Implement Whisper-based audio transcription tool with Gradio UI; update README and add requirements