- Feb 12, 2025
-
-
Mathieu Duponchelle authored
Part-of: <gstreamer/gst-plugins-rs!2071>
-
Guillaume Desmottes authored
Wait for each stream to EOS rather than rely on the EOS message. Prevent races as our test files are really short, see gstreamer/gstreamer#4204 Part-of: <gstreamer/gst-plugins-rs!2063>
-
- Feb 11, 2025
-
-
Part-of: <!2069>
-
As stated in the spec: > Bit b7 of the LINE value is the field number (0 for field 2; > 1 for field 1). Bits b6 and b5 are 0. Bits b4-b0 form a 5-bit unsigned > integer which represents the offset [..] Here, b7 is the most significant bit, this is what both cccombiner and ccconverter agree on. The overlay elements previously looked at the least significant bit, which led to garbled text when both fields were actually present in the CC meta. Part-of: <!2069>
-
- Feb 10, 2025
-
-
Part-of: <!2055>
-
The original awstranscribe element has grown too complex when integrating translations for reasons that in retrospect were wrong: As awstranscribe outputs words one by one, I decided we wanted to perform translations there with larger sentences if available, but an alternative design where a separate translation element is composed downstream is also possible, as long as that element accumulates words and enough latency is set on the transcriber. An important difference is that the new elements do not expose unsynced pads, this use case is instead now served by simple messages on the bus. The elements should otherwise be at feature parity with the original element. A higher-level bin is also provided for convenience (and usage within transcriberbin): translationbin. A transcriber element can be provided to this bin, which exposes an always audio sink pad, and an always text sink pad (for the transcripts). Additional source pads can ...
-
This improves readability with speaker-detection capable elements upstream. Part-of: <!2055>
-
diarization=speaker can be set to enable speaker detection, and max-speakers can be set to control the maximum number of detected speakers. An event is then forwarded downstream upon speaker changes. Part-of: <!2055>
-
This deprecates the buffers pushed on the unsynced pads, which should be removed prior to release. Part-of: <!2055>
-
There is no reason to delay the output of items until the deadline. Part-of: <!2055>
-
This allows controlling the requested delay independently from the latency of the element. Part-of: <!2055>
-
- Feb 07, 2025
-
-
Guillaume Desmottes authored
Sending a serialized query may potentially block so best to release locks before doing so. Hopefully fix a dead lock I analyzed from a post mortem report. Part-of: <!2060>
-
- Feb 03, 2025
-
-
Currently, the typefind logic for cdg can cause errors on files with all conditions below: - size multiple of NB_WINDOWS * CDG_PACKET_SIZE (8 * 24) = 192 - size less than TYPEFIND_SEARCH_WINDOW (28800) - extension different than .cdg With those conditions, the cdg typefind function may be the last one to execute and cause gsttypefindhelper.c::helper_find_peek to return GST_FLOW_EOS, and if it's the final call will then be turned to GST_FLOW_ERROR in typefind. That is because cdg's typefind will perform a last call to typefind.peek(len, CDG_PACKET_SIZE) which is outside the range of typefind (from 0 to len - 1). Reproducer with file from cdg tests: ``` head -n 28608 BrotherJohn.cdg > BrotherJohn_short.bit gst-play-1.0 BrotherJohn_short.bit ``` Fix this by correcting the loop bounds in the cdg typefind logic. Perhaps the typefind helper logic should be rewritten to avoid such cases too. Then the example above works and unrelated files aren't affected by this in typefind anymore. Part-of: <!2056>
-
- Jan 31, 2025
-
-
Sanchayan Maity authored
Part-of: <!2054>
-
Sanchayan Maity authored
This copies over the required bits of code from cros-codecs and drops the dependency on cros-codecs. Part-of: <!2054>
-
- Jan 30, 2025
-
-
Sanchayan Maity authored
Part-of: <!2054>
-
- Jan 28, 2025
-
-
The row used in roll up mode was inconsistently referenced (row vs base_row) and could result in incorrect output being produced. Part-of: <!2051>
-
Part-of: <!2051>
-
Otherwise, when iterating over the lines, they may be written out of order. Part-of: <!2051>
-
Sanchayan Maity authored
Part-of: <!1515>
-
Sanchayan Maity authored
This is required for the hlsmultivariantsink tests to work. Also register the plugins before running the test. Part-of: <!1515>
-
Sanchayan Maity authored
For the MPEG-TS case, we depend on cros-codecs for parsing SPS to get the relevant information for building codec strings. Do not compile it for non-linux platforms. Users needing MPEG-TS on non-linux platform need to set codec string manually as a workaround. Part-of: <!1515>
-
Sanchayan Maity authored
Part-of: <!1515>
-
Sanchayan Maity authored
`hlsmultivariantsink` adds support for the following as per RFC 8216 - Multivariant/master playlist - Alternate Renditions - Variant Streams Part-of: <!1515>
-
Yaakov Selkowitz authored
This should fix the crates that are missing license files. Part-of: <!2050>
-
- Jan 21, 2025
-
-
Sebastian Dröge authored
Simply skip over them instead of handling them as parse error. Part-of: <!2047>
-
This adds support for direct encoding of common formats into ISO base media file format. There are unit tests for formats that are not completely supported, to check that those functions work correctly, and to ease future extension. End-to-end testing currently requires use of gpac to validate files. Part-of: <!1990>
-
- Jan 20, 2025
-
-
Check for valid frame header. Part-of: <!2044>
-
So we can actually check the content of depayloaded buffers too. Part-of: <!2044>
-
The frames of a non-fragmented payload would contain an extra two bytes before the frame sync and then missing two bytes at the end which which would cause decoding errors on the last block and/or frame crc check failures. This happened because we didn't take into account the 2-byte packet payload header when creating output sub-buffers, as the offsets we were using were in relation to the payload data after the headers. Fixes #645 Part-of: <!2044>
-
This way an application can tell the language code right away at pad-added time. Part-of: <!2029>
-
In order to do so we need to activate the pad as soon as it is added, which means we can no longer start the task at this point, instead wait for stream-start to do so now. Part-of: <!2029>
-
Sebastian Dröge authored
Part-of: <!2045>
-
Sebastian Dröge authored
Part-of: <!2045>
-
- Jan 16, 2025
-
-
Update both tutorial-1.md and tutorial-2.md to match the changes done in the tutorials implementation regarding Lazy to LazyLock Part-of: <!2041>
-
- Jan 15, 2025
-
-
If there are pending buffers without a segment then they must come from the caps only and should be forwarded at a later time, if any. Also reject any incoming buffers if no segment was received. Part-of: <!2037>
-
If there are pending buffers without a segment then they must come from the caps only and should be forwarded at a later time, if any. Also reject any incoming buffers if no segment was received. Part-of: <!2037>
-
- Jan 14, 2025
-
-
Brings support for multiple streams of each kind to fallbacksrc. Usage past 1video/1audio stream now requires using the stream selection API. fallbacksrc will expose its own collection of streams, which will be mapped to streams from the main and fallback source automatically. This mapping can be changed via the map-streams signal. The amount of streams being exposed by fallbacksrc is dictated by the main source. CustomSource has been updated to also support multi-stream scenarios, both for stream-aware elements and for simple bins without such functionality. Co-authored-by:
Sebastian Dröge <sebastian@centricular.com> Part-of: <!1832>
-
Sebastian Dröge authored
Part-of: <!2040>
-
Sebastian Dröge authored
It requires Rust 1.81 or newer. Part-of: <!2040>
-