PocketSphinx ps_process_raw sample buffer size effect
.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty{ height:90px;width:728px;box-sizing:border-box;
}
In pocketsphinx tutorial, the ps_process_raw function was called with a buffer of 512 int16s:
int16 buf[512];
while (!feof(fh)) {
size_t nsamp;
nsamp = fread(buf, 2, 512, fh);
ps_process_raw(ps, buf, nsamp, FALSE, FALSE);
}
However in their own source code called by the command line interface, they used 256:
total = 0;
while (!feof(rawfh)) {
int16 data[256];
size_t nread;
nread = fread(data, sizeof(*data), sizeof(data)/sizeof(*data), rawfh);
ps_process_raw(ps, data, nread, FALSE, FALSE);
total += nread;
}
Elsewhere I've seen 2048 as well. If I understand correctly, given input audio is preferred in 16khz, 512 samples corresponds to around 30 ms of sound, which lines up with what most speech recognition guides seem to suggest.
I'm wondering if anyone knows how sample buffer size affects performance (accuracy and speed), and why it isn't consistent across different 'official' sources.
I'm on the latest sphinx version using the C++ API.
c++ cmusphinx pocketsphinx
add a comment |
In pocketsphinx tutorial, the ps_process_raw function was called with a buffer of 512 int16s:
int16 buf[512];
while (!feof(fh)) {
size_t nsamp;
nsamp = fread(buf, 2, 512, fh);
ps_process_raw(ps, buf, nsamp, FALSE, FALSE);
}
However in their own source code called by the command line interface, they used 256:
total = 0;
while (!feof(rawfh)) {
int16 data[256];
size_t nread;
nread = fread(data, sizeof(*data), sizeof(data)/sizeof(*data), rawfh);
ps_process_raw(ps, data, nread, FALSE, FALSE);
total += nread;
}
Elsewhere I've seen 2048 as well. If I understand correctly, given input audio is preferred in 16khz, 512 samples corresponds to around 30 ms of sound, which lines up with what most speech recognition guides seem to suggest.
I'm wondering if anyone knows how sample buffer size affects performance (accuracy and speed), and why it isn't consistent across different 'official' sources.
I'm on the latest sphinx version using the C++ API.
c++ cmusphinx pocketsphinx
If I understand this correctly the size of your buffer doesn't really impact accuracy. It could inpact speed though. Larger buffers read more of a file at a time and in my perfessional experience it's faster to load files directly into memory then process them rather than processing them as you read them from disk to memory. What does affect accuracy is your file format and the tutorial is extremely specific about what format the file needs to be in if you was to create your own.
– johnathan
Nov 17 '18 at 1:21
@johnathon I see, thanks. I'll play around and see how results differ.
– NoviceEngineer
Nov 17 '18 at 6:34
add a comment |
In pocketsphinx tutorial, the ps_process_raw function was called with a buffer of 512 int16s:
int16 buf[512];
while (!feof(fh)) {
size_t nsamp;
nsamp = fread(buf, 2, 512, fh);
ps_process_raw(ps, buf, nsamp, FALSE, FALSE);
}
However in their own source code called by the command line interface, they used 256:
total = 0;
while (!feof(rawfh)) {
int16 data[256];
size_t nread;
nread = fread(data, sizeof(*data), sizeof(data)/sizeof(*data), rawfh);
ps_process_raw(ps, data, nread, FALSE, FALSE);
total += nread;
}
Elsewhere I've seen 2048 as well. If I understand correctly, given input audio is preferred in 16khz, 512 samples corresponds to around 30 ms of sound, which lines up with what most speech recognition guides seem to suggest.
I'm wondering if anyone knows how sample buffer size affects performance (accuracy and speed), and why it isn't consistent across different 'official' sources.
I'm on the latest sphinx version using the C++ API.
c++ cmusphinx pocketsphinx
In pocketsphinx tutorial, the ps_process_raw function was called with a buffer of 512 int16s:
int16 buf[512];
while (!feof(fh)) {
size_t nsamp;
nsamp = fread(buf, 2, 512, fh);
ps_process_raw(ps, buf, nsamp, FALSE, FALSE);
}
However in their own source code called by the command line interface, they used 256:
total = 0;
while (!feof(rawfh)) {
int16 data[256];
size_t nread;
nread = fread(data, sizeof(*data), sizeof(data)/sizeof(*data), rawfh);
ps_process_raw(ps, data, nread, FALSE, FALSE);
total += nread;
}
Elsewhere I've seen 2048 as well. If I understand correctly, given input audio is preferred in 16khz, 512 samples corresponds to around 30 ms of sound, which lines up with what most speech recognition guides seem to suggest.
I'm wondering if anyone knows how sample buffer size affects performance (accuracy and speed), and why it isn't consistent across different 'official' sources.
I'm on the latest sphinx version using the C++ API.
c++ cmusphinx pocketsphinx
c++ cmusphinx pocketsphinx
asked Nov 16 '18 at 21:47
NoviceEngineerNoviceEngineer
105
105
If I understand this correctly the size of your buffer doesn't really impact accuracy. It could inpact speed though. Larger buffers read more of a file at a time and in my perfessional experience it's faster to load files directly into memory then process them rather than processing them as you read them from disk to memory. What does affect accuracy is your file format and the tutorial is extremely specific about what format the file needs to be in if you was to create your own.
– johnathan
Nov 17 '18 at 1:21
@johnathon I see, thanks. I'll play around and see how results differ.
– NoviceEngineer
Nov 17 '18 at 6:34
add a comment |
If I understand this correctly the size of your buffer doesn't really impact accuracy. It could inpact speed though. Larger buffers read more of a file at a time and in my perfessional experience it's faster to load files directly into memory then process them rather than processing them as you read them from disk to memory. What does affect accuracy is your file format and the tutorial is extremely specific about what format the file needs to be in if you was to create your own.
– johnathan
Nov 17 '18 at 1:21
@johnathon I see, thanks. I'll play around and see how results differ.
– NoviceEngineer
Nov 17 '18 at 6:34
If I understand this correctly the size of your buffer doesn't really impact accuracy. It could inpact speed though. Larger buffers read more of a file at a time and in my perfessional experience it's faster to load files directly into memory then process them rather than processing them as you read them from disk to memory. What does affect accuracy is your file format and the tutorial is extremely specific about what format the file needs to be in if you was to create your own.
– johnathan
Nov 17 '18 at 1:21
If I understand this correctly the size of your buffer doesn't really impact accuracy. It could inpact speed though. Larger buffers read more of a file at a time and in my perfessional experience it's faster to load files directly into memory then process them rather than processing them as you read them from disk to memory. What does affect accuracy is your file format and the tutorial is extremely specific about what format the file needs to be in if you was to create your own.
– johnathan
Nov 17 '18 at 1:21
@johnathon I see, thanks. I'll play around and see how results differ.
– NoviceEngineer
Nov 17 '18 at 6:34
@johnathon I see, thanks. I'll play around and see how results differ.
– NoviceEngineer
Nov 17 '18 at 6:34
add a comment |
0
active
oldest
votes
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53345882%2fpocketsphinx-ps-process-raw-sample-buffer-size-effect%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
0
active
oldest
votes
0
active
oldest
votes
active
oldest
votes
active
oldest
votes
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53345882%2fpocketsphinx-ps-process-raw-sample-buffer-size-effect%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
If I understand this correctly the size of your buffer doesn't really impact accuracy. It could inpact speed though. Larger buffers read more of a file at a time and in my perfessional experience it's faster to load files directly into memory then process them rather than processing them as you read them from disk to memory. What does affect accuracy is your file format and the tutorial is extremely specific about what format the file needs to be in if you was to create your own.
– johnathan
Nov 17 '18 at 1:21
@johnathon I see, thanks. I'll play around and see how results differ.
– NoviceEngineer
Nov 17 '18 at 6:34