Skip to content

Commit 4281a5d

Browse files
RCmagsRCmags
committed
change file order
1 parent 5830303 commit 4281a5d

File tree

4 files changed

+32
-18
lines changed

4 files changed

+32
-18
lines changed

‎collections/blog/Journal/_posts/2025-03-10-dictation-writting.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -13,7 +13,7 @@ On my [ThinkPad T540P](https://static.lenovo.com/shop/emea/content/pdf/ThinkPad/
1313

1414
When combined with other AI tools like Large Language Models (e.g. [Mistral:7b](https://ollama.com/library/mistral) or [DeepSeek-R1:8b](https://ollama.com/library/deepseek-r1) ), these transcribed texts can be refined significantly. For instance, pairing SpeechNote's transcription with a local LLM service like [Ollama](https://ollama.com/) allows the LLM to correct spelling and improve sentence structure. This results in a fairly refined piece of text. Therefore, both tools complement each other well: we transcribe thoughts from speech, then use an LLM model to polish that the transcribed text into something clearer and more readable.
1515

16-
![image](/img/ai-dictation/ai-speech-correction.png)
16+
![image](/diagrams/ai-dictation/ai-speech-correction.png)
1717
**File**: [ai-speech-correction.drawio](/diagrams/ai-speech-correction.drawio)
1818

1919
This refined text can be reviewed and edited by the user for final touches, making it closer to their intended expression. Critically, this approach drastically reduces the manual effort required compared to writing from scratch. Moreover, I believe speaking naturally leverages neural pathways that are optimized for rapid idea generation. Speech is our primary mode of communication and it forces us to express clear ideas quickly. Writing might involve over-refining thoughts that would flow more freely during speech. Transcription appears to bypass these internal filters and makes communicating an idea much more natural. The speed increase is substantial: dictation yields far more words per minute than manual typing ever can. It only requires an accurate transcription model and a sufficiently powerful computer.

diagrams/ai-speech-correction.drawio renamed to diagrams/ai-dictation/ai-speech-correction.drawio

Lines changed: 31 additions & 17 deletions
Original file line numberDiff line numberDiff line change
@@ -1,61 +1,75 @@
11
<mxfile host="Electron" agent="Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) draw.io/26.1.1 Chrome/132.0.6834.210 Electron/34.3.3 Safari/537.36" version="26.1.1">
22
<diagram name="Page-1" id="PLBad71tbFdv3gAetYYm">
3-
<mxGraphModel dx="741" dy="426" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="1100" pageHeight="850" math="0" shadow="0">
3+
<mxGraphModel dx="894" dy="514" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="1100" pageHeight="850" math="0" shadow="0">
44
<root>
55
<mxCell id="0" />
66
<mxCell id="1" parent="0" />
7-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-2" value="" style="html=1;strokeWidth=1;shadow=0;dashed=0;shape=mxgraph.ios7.misc.text_size;fillColor=#007AFF;strokeColor=none;buttonText=;strokeColor2=#222222;fontColor=#222222;fontSize=8;verticalLabelPosition=bottom;verticalAlign=top;align=center;sketch=0;" vertex="1" parent="1">
7+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-2" value="" style="html=1;strokeWidth=1;shadow=0;dashed=0;shape=mxgraph.ios7.misc.text_size;fillColor=#007AFF;strokeColor=none;buttonText=;strokeColor2=#222222;fontColor=#222222;fontSize=8;verticalLabelPosition=bottom;verticalAlign=top;align=center;sketch=0;" parent="1" vertex="1">
88
<mxGeometry x="430" y="258.5" width="60" height="57.5" as="geometry" />
99
</mxCell>
10-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-4" value="" style="image;aspect=fixed;html=1;points=[];align=center;fontSize=12;image=img/lib/azure2/ai_machine_learning/Speech_Services.svg;" vertex="1" parent="1">
10+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-4" value="" style="image;aspect=fixed;html=1;points=[];align=center;fontSize=12;image=img/lib/azure2/ai_machine_learning/Speech_Services.svg;" parent="1" vertex="1">
1111
<mxGeometry x="290" y="253.5" width="67.5" height="67.5" as="geometry" />
1212
</mxCell>
13-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="-A8UZ74H-OdVrFeQKU_n-5" target="-A8UZ74H-OdVrFeQKU_n-6">
13+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" parent="1" source="-A8UZ74H-OdVrFeQKU_n-5" target="-A8UZ74H-OdVrFeQKU_n-6" edge="1">
1414
<mxGeometry relative="1" as="geometry" />
1515
</mxCell>
16-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-5" value="" style="image;sketch=0;aspect=fixed;html=1;points=[];align=center;fontSize=12;image=img/lib/mscae/Cognitive_Services.svg;" vertex="1" parent="1">
16+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-5" value="" style="image;sketch=0;aspect=fixed;html=1;points=[];align=center;fontSize=12;image=img/lib/mscae/Cognitive_Services.svg;" parent="1" vertex="1">
1717
<mxGeometry x="550" y="258.5" width="95.83" height="57.5" as="geometry" />
1818
</mxCell>
19-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="-A8UZ74H-OdVrFeQKU_n-6">
19+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" parent="1" source="-A8UZ74H-OdVrFeQKU_n-6" edge="1">
2020
<mxGeometry relative="1" as="geometry">
2121
<mxPoint x="740" y="360" as="targetPoint" />
2222
</mxGeometry>
2323
</mxCell>
24-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-6" value="" style="sketch=0;pointerEvents=1;shadow=0;dashed=0;html=1;strokeColor=none;fillColor=#505050;labelPosition=center;verticalLabelPosition=bottom;verticalAlign=top;outlineConnect=0;align=center;shape=mxgraph.office.concepts.address_book;" vertex="1" parent="1">
24+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-6" value="" style="sketch=0;pointerEvents=1;shadow=0;dashed=0;html=1;strokeColor=none;fillColor=#505050;labelPosition=center;verticalLabelPosition=bottom;verticalAlign=top;outlineConnect=0;align=center;shape=mxgraph.office.concepts.address_book;" parent="1" vertex="1">
2525
<mxGeometry x="700" y="256.56" width="80" height="59.88" as="geometry" />
2626
</mxCell>
27-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0;entryY=0.5;entryDx=0;entryDy=0;entryPerimeter=0;" edge="1" parent="1" source="-A8UZ74H-OdVrFeQKU_n-2" target="-A8UZ74H-OdVrFeQKU_n-5">
27+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0;entryY=0.5;entryDx=0;entryDy=0;entryPerimeter=0;" parent="1" source="-A8UZ74H-OdVrFeQKU_n-2" target="-A8UZ74H-OdVrFeQKU_n-5" edge="1">
2828
<mxGeometry relative="1" as="geometry" />
2929
</mxCell>
30-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0.01;entryY=0.493;entryDx=0;entryDy=0;entryPerimeter=0;" edge="1" parent="1" source="-A8UZ74H-OdVrFeQKU_n-4" target="-A8UZ74H-OdVrFeQKU_n-2">
30+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0.01;entryY=0.493;entryDx=0;entryDy=0;entryPerimeter=0;" parent="1" source="-A8UZ74H-OdVrFeQKU_n-4" target="-A8UZ74H-OdVrFeQKU_n-2" edge="1">
3131
<mxGeometry relative="1" as="geometry" />
3232
</mxCell>
33-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-15" value="Speech-to-Text Model" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
33+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-15" value="Speech-to-Text Model" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" parent="1" vertex="1">
3434
<mxGeometry x="277.5" y="216.5" width="92.5" height="30" as="geometry" />
3535
</mxCell>
36-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-16" value="LLM model" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
36+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-16" value="LLM model" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" parent="1" vertex="1">
3737
<mxGeometry x="557.92" y="223.5" width="80" height="30" as="geometry" />
3838
</mxCell>
39-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-17" value="Transcribed text" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
39+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-17" value="Transcribed text" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" parent="1" vertex="1">
4040
<mxGeometry x="411.25" y="223.5" width="97.5" height="30" as="geometry" />
4141
</mxCell>
42-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-18" value="Refined Draft" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
42+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-18" value="Refined Draft" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" parent="1" vertex="1">
4343
<mxGeometry x="690" y="223.5" width="100" height="30" as="geometry" />
4444
</mxCell>
45-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-22" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" target="-A8UZ74H-OdVrFeQKU_n-5">
45+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-22" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" parent="1" target="-A8UZ74H-OdVrFeQKU_n-5" edge="1">
4646
<mxGeometry relative="1" as="geometry">
4747
<mxPoint x="710" y="405.02857142857135" as="sourcePoint" />
4848
</mxGeometry>
4949
</mxCell>
50-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-20" value="" style="verticalLabelPosition=bottom;html=1;verticalAlign=top;align=center;strokeColor=none;fillColor=#00BEF2;shape=mxgraph.azure.user;" vertex="1" parent="1">
50+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-20" value="" style="verticalLabelPosition=bottom;html=1;verticalAlign=top;align=center;strokeColor=none;fillColor=#00BEF2;shape=mxgraph.azure.user;" parent="1" vertex="1">
5151
<mxGeometry x="716.25" y="370" width="47.5" height="50" as="geometry" />
5252
</mxCell>
53-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-23" value="Human Review" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
53+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-23" value="Human Review" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" parent="1" vertex="1">
5454
<mxGeometry x="770" y="390" width="60" height="30" as="geometry" />
5555
</mxCell>
56-
<mxCell id="-A8UZ74H-OdVrFeQKU_n-25" value="Request corrections" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#999999;" vertex="1" parent="1">
56+
<mxCell id="-A8UZ74H-OdVrFeQKU_n-25" value="Request corrections" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#999999;" parent="1" vertex="1">
5757
<mxGeometry x="590" y="370" width="92.08" height="30" as="geometry" />
5858
</mxCell>
59+
<mxCell id="MM8X_Phhc6biQTYKt19l-1" value="" style="html=1;verticalLabelPosition=bottom;align=center;labelBackgroundColor=#ffffff;verticalAlign=top;strokeWidth=2;strokeColor=#0080F0;shadow=0;dashed=0;shape=mxgraph.ios7.icons.microphone;" vertex="1" parent="1">
60+
<mxGeometry x="181" y="253.5" width="30" height="62" as="geometry" />
61+
</mxCell>
62+
<mxCell id="MM8X_Phhc6biQTYKt19l-4" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=-0.035;entryY=0.47;entryDx=0;entryDy=0;entryPerimeter=0;" edge="1" parent="1" target="-A8UZ74H-OdVrFeQKU_n-4">
63+
<mxGeometry relative="1" as="geometry">
64+
<mxPoint x="220" y="285" as="sourcePoint" />
65+
<Array as="points">
66+
<mxPoint x="260" y="286" />
67+
</Array>
68+
</mxGeometry>
69+
</mxCell>
70+
<mxCell id="MM8X_Phhc6biQTYKt19l-6" value="User Speaks ideas" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#999999;" vertex="1" parent="1">
71+
<mxGeometry x="213" y="300" width="70" height="30" as="geometry" />
72+
</mxCell>
5973
</root>
6074
</mxGraphModel>
6175
</diagram>

‎diagrams/ai-dictation/ai-speech-correction.png‎

36.5 KB
Loading

‎img/ai-dictation/ai-speech-correction.png‎

-30.8 KB
Binary file not shown.

0 commit comments

Comments
 (0)