Skip to content
Snippets Groups Projects
Commit 7a3a3e78 authored by Ahmet Oner's avatar Ahmet Oner
Browse files

Upgrade OpenAI Whisper to v20231106 (large-v3)

parent f9ca4dc1
No related branches found
No related tags found
No related merge requests found
......@@ -4,6 +4,12 @@ Changelog
Unreleased
----------
### Changed
- Upgraded
- [openai/whisper](https://github.com/openai/whisper) to [v20231106](https://github.com/openai/whisper/releases/tag/v20231106)
[1.2.2] (2023-11-03)
--------------------
......
......@@ -9,7 +9,7 @@ Whisper is a general-purpose speech recognition model. It is trained on a large
## Features
Current release (v1.2.2) supports following whisper models:
- [openai/whisper](https://github.com/openai/whisper)@[v20230918](https://github.com/openai/whisper/releases/tag/v20230918)
- [openai/whisper](https://github.com/openai/whisper)@[v20231106](https://github.com/openai/whisper/releases/tag/v20231106)
- [guillaumekln/faster-whisper](https://github.com/guillaumekln/faster-whisper)@[0.9.0](https://github.com/guillaumekln/faster-whisper/releases/tag/v0.9.0)
......
......@@ -15,7 +15,7 @@
export ASR_MODEL=base
```
Available ASR_MODELs are `tiny`, `base`, `small`, `medium`, `large` (only OpenAI Whisper), `large-v1` and `large-v2`. Please note that `large` and `large-v2` are the same model.
Available ASR_MODELs are `tiny`, `base`, `small`, `medium`, `large` (only OpenAI Whisper), `large-v1`, `large-v2` and `large-v3` (only OpenAI Whisper for now).
For English-only applications, the `.en` models tend to perform better, especially for the `tiny.en` and `base.en` models. We observed that the difference becomes less significant for the `small.en` and `medium.en` models.
......
......@@ -3,7 +3,7 @@ Whisper is a general-purpose speech recognition model. It is trained on a large
## Features
Current release (v1.2.2) supports following whisper models:
- [openai/whisper](https://github.com/openai/whisper)@[v20230918](https://github.com/openai/whisper/releases/tag/v20230918)
- [openai/whisper](https://github.com/openai/whisper)@[v20231106](https://github.com/openai/whisper/releases/tag/v20231106)
- [guillaumekln/faster-whisper](https://github.com/guillaumekln/faster-whisper)@[0.9.0](https://github.com/guillaumekln/faster-whisper/releases/tag/v0.9.0)
## Quick Usage
......
......@@ -757,19 +757,19 @@ sympy = "*"
[[package]]
name = "openai-whisper"
version = "20230918"
version = "20231106"
description = "Robust Speech Recognition via Large-Scale Weak Supervision"
optional = false
python-versions = ">=3.8"
files = [
{file = "openai-whisper-20230918.tar.gz", hash = "sha256:32a1ee39c3faaf6c719e3a83f1aacc8e164aad87976350371e26845271287c30"},
{file = "openai-whisper-20231106.tar.gz", hash = "sha256:9d1de7fa1e766b9adf8be4bfa7fb11e2bdf8d2b0bf77b90478cf4d75e0e58d19"},
]
[package.dependencies]
more-itertools = "*"
numba = "*"
numpy = "*"
tiktoken = "0.3.3"
tiktoken = "*"
torch = "*"
tqdm = "*"
......@@ -1574,4 +1574,4 @@ files = [
[metadata]
lock-version = "2.0"
python-versions = "^3.10"
content-hash = "7ba666ad744fc80fff978862f014431c8a03565c7d1b6a459bc759ace4f73078"
content-hash = "d4773fb3718894fe55d566f19e641fd7bd9be1516788eca099c32cd10c4bd0df"
......@@ -26,7 +26,7 @@ ffmpeg-python = "^0.2.0"
fastapi = "^0.104.1"
llvmlite = "^0.41.1"
numba = "^0.58.0"
openai-whisper = "20230918"
openai-whisper = "20231106"
faster-whisper = "^0.9.0"
torch = [
{markers = "sys_platform == 'darwin' and platform_machine == 'arm64'", url = "https://download.pytorch.org/whl/cpu/torch-1.13.0-cp310-none-macosx_11_0_arm64.whl"},
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment