Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/kushal211/speechrecognizer
A SwiftUI-based iOS application that uses Apple’s Speech framework to perform real-time speech-to-text transcription. This app supports starting, pausing, resuming, and stopping speech recognition, with an optional auto-stop feature based on a configurable silence duration.
https://github.com/kushal211/speechrecognizer
ios speech-to-text speechrecognizer swiftui visionos
Last synced: 11 days ago
JSON representation
A SwiftUI-based iOS application that uses Apple’s Speech framework to perform real-time speech-to-text transcription. This app supports starting, pausing, resuming, and stopping speech recognition, with an optional auto-stop feature based on a configurable silence duration.
- Host: GitHub
- URL: https://github.com/kushal211/speechrecognizer
- Owner: kushal211
- License: mit
- Created: 2024-12-12T10:22:24.000Z (11 days ago)
- Default Branch: main
- Last Pushed: 2024-12-12T11:13:53.000Z (11 days ago)
- Last Synced: 2024-12-12T11:29:31.136Z (11 days ago)
- Topics: ios, speech-to-text, speechrecognizer, swiftui, visionos
- Language: Swift
- Homepage:
- Size: 226 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# SpeechRecognizer
A SwiftUI-based iOS application that uses Apple’s Speech framework to perform real-time speech-to-text transcription. This app supports starting, pausing, resuming, and stopping speech recognition, with an optional auto-stop feature based on a configurable silence duration.
---
### Features
* Real-time speech-to-text transcription.
* Start, Pause, Resume, and Stop speech recognition.
* Optional auto-stop functionality triggered after a specified duration of silence.
* Configurable silence duration for flexible behavior.
* Logs the final recognized text to the console when recognition stops.
* Dynamic UI reflecting the current recognition state.
---
### Requirements
* iOS 14.0+
* Xcode 15.0+
* Swift 5.0+
* Device with microphone access and speech recognition permissions enabled.
---
### Installation
1. Clone the repository:
```
git clone
```
2. Open the project in Xcode:
```
cd
open SpeechRecognizer.xcodeproj
```
1. Build and run the project on a real iOS device or simulator (microphone access required).
---
### Usage
1. Launch the app.
2. Start Speech Recognition:
* Tap the green “**Play**” button to begin speech recognition.
* Speak into the microphone to see real-time transcription.
3. Pause Speech Recognition:
* Tap the orange “**Pause**” button to temporarily halt recognition.
4. Resume Speech Recognition:
* Tap the green “**Play**” button again to continue recognition.
5. Stop Speech Recognition:
* Tap the red “**Stop**” button to stop recognition and log the final transcribed text.
##### Auto-Stop Feature
* By default, the app will automatically stop recognition after 5 seconds of silence.
* To customize or disable this feature:
* Set `silenceDuration` in SpeechAnalyzer:
```
speechAnalyzer.silenceDuration = 10 // Set to 10 seconds
speechAnalyzer.silenceDuration = 0 // Disable auto-stop
```
---
### Code Overview
**SpeechAnalyzer**
* **Purpose**:
* Core logic for managing speech recognition using SFSpeechRecognizer and AVAudioEngine.
* **Key Methods**:
* `start()`: Starts speech recognition.
* `pause()`: Pauses the recognition process.
* `resume()`: Resumes the paused recognition.
* `stop()`: Stops recognition and logs the final result.
* **Auto-Stop Behavior**:
* Monitors silence using a configurable silenceDuration.
* Stops recognition automatically if no speech is detected within the specified duration.
* **SpeechRecognitionView**
* Purpose: Provides a user interface to interact with the SpeechAnalyzer.
* Key Features:
* Buttons for starting, pausing, resuming, and stopping speech recognition.
* Dynamic button states based on recognition status.
* Displays the recognized text in real time.
---
### Permissions
The app requires the following permissions:
* *Microphone Access*: To record audio for speech recognition.
* *Speech Recognition* Access: To perform transcription.
Ensure permissions are granted in your app’s `Info.plist`:
```
NSMicrophoneUsageDescription
We need access to your microphone for speech recognition.
NSSpeechRecognitionUsageDescription
We need access to speech recognition for transcribing your speech.
```
---
### Customization
1. UI Customization:
* Modify SpeechRecognitionView to change button styles, text appearance, or layout.
2. Silence Duration:
* Update silenceDuration in SpeechAnalyzer for custom auto-stop behavior.
---
### Troubleshooting
* Speech Recognition Fails to Start:
* Ensure permissions are granted for microphone and speech recognition.
* Test on a real device if simulator doesn’t support speech recognition.
* Auto-Stop Not Working:
* Verify that silenceDuration is greater than 0.
---
### License
This project is licensed under the MIT License. See the LICENSE file for details.
---
### Acknowledgments
This app leverages the following Apple frameworks:
* `Speech Framework`: For speech-to-text transcription.
* `AVFoundation`: For audio input and processing.
---
### Contact
[Kushal Panchal][portray-url]
[![LinkedIn][linkedin-shield]][linkedin-url]
[linkedin-shield]: https://img.shields.io/badge/LinkedIn-0077B5?style=for-the-badge&logo=linkedin&logoColor=white
[portray-url]: https://www.portray.work/kushal
[linkedin-url]: https://www.linkedin.com/in/kushal211