Accept 16kHz S16 audio buffers on its sink pad and produce text buffers on its source pad.

gst-whisper is written in Rust and uses the cargo-c helper. Set up a Rust development environment (e.g. using rustup) and then:

cargo install cargo-c

git clone https://github.com/avstack/gst-whisper
cd gst-whisper
cargo cbuild --release
export GST_PLUGIN_PATH=$(pwd)/target/x86_64-apple-darwin/release

You must already have the Whisper model.

export WHISPER_MODEL_PATH=../whisper.cpp/models/ggml-base.en.bin

gst-launch-1.0 filesrc location=../whisper.cpp/samples/jfk.wav ! decodebin ! audioconvert ! queue ! whisper ! fdsink

gst-whisper is licensed under either of

• Apache License, Version 2.0, (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
• MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.

Any kinds of contributions are welcome as a pull request.

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in these crates by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.

gst-meet development is sponsored by AVStack. We provide globally-distributed, scalable, managed Jitsi Meet backends.