faster-whisper is a reimplementation of OpenAI's Whisper model using
CTranslate2, which is up to 4 times faster than enai/whisper for the
same accuracy while using less memory. The efficiency can be further
improved with 8-bit quantization on both CPU and GPU.
It can automatically detect the following 14 languages and transcribe
the text into their respective languages: en, zh, fr, de, ja, ko, ru,
es, th, it, pt, vi, ar, tr.
The gitbub repository for faster-whisper is :
https://github.com/SYSTRAN/faster-whisper
---------
Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
**Description:** The base URL for OpenAI is retrieved from the
environment variable "OPENAI_BASE_URL", whereas for langchain it is
obtained from "OPENAI_API_BASE". By adding `base_url =
os.environ.get("OPENAI_API_BASE")`, the OpenAI proxy can execute
correctly.
---------
Co-authored-by: Bagatur <baskaryan@gmail.com>
**Description**
This PR addresses a rare issue in `OpenAIWhisperParser` that causes it
to crash when processing an audio file with a duration very close to the
class's chunk size threshold of 20 minutes.
**Issue**
#11449
**Dependencies**
None
**Tag maintainer**
@agola11 @eyurtsev
**Twitter handle**
leonardodiegues
---------
Co-authored-by: Leonardo Diegues <leonardo.diegues@grupofolha.com.br>
Co-authored-by: Bagatur <baskaryan@gmail.com>
This PR addresses an issue in OpenAIWhisperParserLocal where requesting
CUDA without availability leads to an AttributeError #15143
Changes:
- Refactored Logic for CUDA Availability: The initialization now
includes a check for CUDA availability. If CUDA is not available, the
code falls back to using the CPU. This ensures seamless operation
without manual intervention.
- Parameterizing Batch Size and Chunk Size: The batch_size and
chunk_size are now configurable parameters, offering greater flexibility
and optimization options based on the specific requirements of the use
case.
---------
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
