:package: package upgrades
Browse files- README.md +3 -3
- requirements.txt +3 -3
README.md
CHANGED
|
@@ -3,11 +3,11 @@ title: Automatic speech recognition
|
|
| 3 |
sdk: gradio
|
| 4 |
app_file: src/app.py
|
| 5 |
python_version: 3.11
|
| 6 |
-
sdk_version: 4.
|
| 7 |
app_port: 7860
|
| 8 |
tags: [asr, stt, speech-to-text, whisper, pyannote, diarization]
|
| 9 |
pinned: true
|
| 10 |
-
emoji:
|
| 11 |
---
|
| 12 |
|
| 13 |
# Automatic speech recognition
|
|
@@ -34,7 +34,7 @@ The repo contains (will contain) code to run the software
|
|
| 34 |
|
| 35 |
### Prerequisites
|
| 36 |
|
| 37 |
-
The host machine must have an
|
| 38 |
|
| 39 |
The graphics card should have at least 12GB VRAM for the largest model.
|
| 40 |
|
|
|
|
| 3 |
sdk: gradio
|
| 4 |
app_file: src/app.py
|
| 5 |
python_version: 3.11
|
| 6 |
+
sdk_version: 4.44.0
|
| 7 |
app_port: 7860
|
| 8 |
tags: [asr, stt, speech-to-text, whisper, pyannote, diarization]
|
| 9 |
pinned: true
|
| 10 |
+
emoji: 👂
|
| 11 |
---
|
| 12 |
|
| 13 |
# Automatic speech recognition
|
|
|
|
| 34 |
|
| 35 |
### Prerequisites
|
| 36 |
|
| 37 |
+
The host machine must have an Nvidia graphics card with CUDA 12.x installed natively, preferably [CUDA 12.1](https://developer.nvidia.com/cuda-12-1-0-download-archive), even when using Docker.
|
| 38 |
|
| 39 |
The graphics card should have at least 12GB VRAM for the largest model.
|
| 40 |
|
requirements.txt
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
-
transformers==4.
|
| 2 |
-
accelerate==0.
|
| 3 |
-
gradio==4.
|
| 4 |
pyannote.audio==3.3.1
|
| 5 |
python-docx==1.1.2
|
| 6 |
pandas==2.2.2
|
|
|
|
| 1 |
+
transformers==4.44.2
|
| 2 |
+
accelerate==0.34.2
|
| 3 |
+
gradio==4.44.0
|
| 4 |
pyannote.audio==3.3.1
|
| 5 |
python-docx==1.1.2
|
| 6 |
pandas==2.2.2
|