-
Notifications
You must be signed in to change notification settings - Fork 32
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
1 changed file
with
11 additions
and
26 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,46 +1,31 @@ | ||
# Integrate Whisper TFLite for Enhanced Mobile App Performance | ||
|
||
If you're interested in incorporating the Whisper TFLite model into your iOS and Android applications, please don't hesitate to reach out to us at [email protected]. Our project features an upgraded iteration of the Whisper quantized TFLite model, finely tuned for optimal performance on both Android and iOS platforms. This model is tailored to excel on edge devices, rendering it versatile for various application scenarios. Contact us for further details and collaboration opportunities | ||
|
||
# Whisper Enhanced Quantized TFLite Model | ||
|
||
This project contains an enhanced version of the Whisper quantized TFLite model optimized for both Android and iOS platforms. The model is designed to perform well on edge devices, making it suitable for a wide range of applications. | ||
|
||
## Requirements | ||
|
||
Before you start using this enhanced Whisper model, you need to install Git Large File Storage (LFS) on your system. This is required because the `whisper-tiny.en.tflite` model used in this project is stored using Git LFS. | ||
|
||
|
||
To install Git LFS (Large File Storage), you can use below command: | ||
|
||
```bash | ||
git lfs install | ||
``` | ||
|
||
Once you've installed Git LFS, you'll be able to work with repositories that use Git LFS for managing large files. | ||
|
||
Remember that you only need to install Git LFS once on your system, and it will work for all Git repositories that use LFS for large file storage. | ||
## Integrate Whisper TFLite for Enhanced Mobile App Performance | ||
|
||
If you're interested in incorporating the Whisper TFLite model into your iOS and Android applications, please don't hesitate to reach out to us at [email protected]. Our project features an upgraded iteration of the Whisper quantized TFLite model, finely tuned for optimal performance on both Android and iOS platforms. This model is tailored to excel on edge devices, rendering it versatile for various application scenarios. Contact us for further details and collaboration opportunities | ||
|
||
For more information, you can follow the [Git LFS installation guide](https://git-lfs.github.com/) to install Git LFS on your system. | ||
|
||
## Getting Started | ||
|
||
To get started with this enhanced Whisper model, follow these steps: | ||
|
||
1. Install Git LFS as mentioned in the requirements section. | ||
|
||
2. Clone this repository to your local machine: | ||
1. Clone this repository to your local machine: | ||
|
||
```bash | ||
git clone https://github.com/nyadla-sys/whisper.tflite.git | ||
``` | ||
|
||
3. You can now use the enhanced Whisper quantized TFLite model in your projects for Android and iOS. | ||
2. You can now use the enhanced Whisper quantized TFLite model in your projects by refering sample code for Android and iOS. | ||
|
||
## Demo Apps | ||
|
||
## Android Example | ||
### Android | ||
You can find a sample Android app in the [whisper_android](whisper_android) folder that demonstrates how to use the Whisper TFLite model for transcription on Android devices. | ||
|
||
You can find a sample Android app in the [android_example](android_example) folder that demonstrates how to use the Whisper TFLite model for transcription on Android devices. | ||
### iOS | ||
You can find a sample iOS app in the [whisper_ios](whisper_ios) folder that demonstrates how to use the Whisper TFLite model for transcription on iOS devices. | ||
|
||
## DTLN quantized tflite model | ||
|
||
|
@@ -54,7 +39,7 @@ Courtesy from [breizhn/DTLN](https://github.com/breizhn/DTLN) | |
|
||
- [ ] Considering adding DTLN noise cancellation tflite model to improve whisper ASR accuracy in noisy environments. | ||
|
||
## Here are some noteworthy links for comparing TFLite with other Whisper models: | ||
## Here are links for comparing TFLite with other Whisper models: | ||
|
||
[Whisper's Comparative Analysis](https://alphacephei.com/nsh/2022/12/11/whisper-other.html) | ||
[Speech Recognition Experiments Repository](https://github.com/fquirin/speech-recognition-experiments) | ||
|