Conversation
v-singh1
commented
Sep 18, 2026
v-singh1
commented
Collaborator
- Publish paired input/output blocks only after reconstructing each chunk. Exclude overlapping context and final padding so concatenated output payloads match the PCM saved to the output WAV.
- Reuse two DMA buffers across synchronous speech processing stages instead of six per-stage allocations. Preserve distinct DSP input/output storage and copy each ISTFT PCM batch before reusing its output buffer.
Speech enhancement sent input PCM before copying audio into the DMA buffer and streamed raw ISTFT output before overlap trimming. This produced zero input payloads and duplicated context in live output visualization. Publish paired input/output blocks only after reconstructing each chunk. Exclude overlapping context and final padding so concatenated output payloads match the PCM saved to the output WAV. - Bound visualization payloads to 1024 samples per message. - Handle odd overlap counts without duplicating boundary samples. - Preserve the final batch when frame counts divide evenly. - Validate overlap bounds and matching STFT/ISTFT time geometry. - Preserve the existing -1 setting for disabled overlap. Signed-off-by: Vishnu Singh <v-singh1@ti.com>
Reuse two DMA buffers across synchronous speech processing stages instead of six per-stage allocations. Preserve distinct DSP input/output storage and copy each ISTFT PCM batch before reusing its output buffer. Add a caller-owned inference buffer overload and use the DMA mapping directly for speech input and output, removing both spectral staging vectors. Preserve the existing vector API and daemon wire protocol. Validate output size before writing the destination and retain CPU-access synchronization across inference. TVM tensor and daemon socket copies remain; this is not end-to-end zero-copy. Signed-off-by: Vishnu Singh <v-singh1@ti.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.